Anda di halaman 1dari 38

The Role of Fuel Preparation in

Low-Emission Combustion
The attainment of very low pollutant emissions, in particular oxides of nitrogen
( NOx), from gas turbines is not only of considerable environmental concern but has
also become an area of increasing competitiveness between the different engine
manufacturers. For stationary engines, the attainment of ultralow NO~ has become
the foremost marketing issue. This paper is devoted primarily to current and emerging
technologies in the development of ultralow emissions combustors for application to
aircraft and stationary engines. Short descriptions of the basic design features of
conventional gas turbine combustors and the methods of fuel injection now in wide-
spread use are followed by a review of fuel spray characteristics and recent develop-
ments in the measurement and modeling of these characteristics. The main gas-
turbine-generated pollutants and their mechanisms of formation are described, along
A. H. Lefebvre with related environmental risks and various issues concerning emissions regulations
Reilly Professor Emeritus of and recently enacted legislation for limiting the pollutant levels emitted by both
Combustion Engineering, aircraft and stationary engines. The impacts of these emissions regulations on com-
Thermal Science and Propulsion Center, bustor and engine design are discussed first in relation to conventional combustors
Purdue UniversLty, and then in the context of variable-geometry and staged combustors. Both these
West Lafayette, IN 47907 concepts are founded on emissions reduction by control of flame temperature. Basic
approaches to the design of "dry" low-NOx and ultralow-NOx combustors are re-
viewed. At the present time lean, premix, prevaporize combustion appears to be
the only technology available for achieving ultralow NOx emissions from practical
combustors. This concept is discussed in some detail, along with its inherent problems
of autoignition, flashback, and acoustic resonance. Attention is also given to alterna-
tive methods of achieving ultralow NOx emissions, notably the rich-burn, quick-
quench, lean-burn, and catalytic combustors. These concepts are now being actively
developed, despite the formidable problems they present in terms of mixing and
durability. The final section reviews the various correlations now being used to
predict the exhaust gas concentrations of the main gaseous pollutant emissions from
gas turbine engines. Comprehensive numerical methods have not yet completely
displaced these semi-empirical correlations but are nevertheless providing useful
insight into the interactions of swirling and recirculating flows with fuel sprays, as
well as guidance to the combustion engineer during the design and development
stages. Throughout the paper emphasis is placed on the important and sometimes
pivotal role played by the fuel preparation process in the reduction of pollutant
emissions from gas turbines.

Introduction tion performance was solved either by modifications to existing

fuel injector designs or by the introduction of new and some-
The gas turbine is often described as the prime mover of the
times revolutionary atomizer concepts. So today, when faced
twentieth century. The tremendous advantages offered by the
with continuous and growing pressure to minimize pollutant
aircraft gas turbine in terms of range, speed, fuel economy, and
emissions from gas turbines, the combustion engineer regards
passenger comfort have assured its preeminent position as the
the fuel preparation process as one that must necessarily play
power plant for aircraft. Also, it continues to find increasingly
a key role in achieving this objective.
wide applications as a power source in a wide variety of indus-
This paper is a review of recent and current research activities
trial and transport applications. In particular, the gas turbine is
in low-emissions gas turbine combustors. It describes research
widely used for both mechanical drive and electrical power
and development work that is directly related to fuel injector
generation within the oil and gas industry. Its advantages in
and combustor design and performance, as opposed to the more
terms of high thermal efficiency and potential for low pollutant
fundamental aspects of atomization, fuel-air mixing, and com-
emissions in the combined cycle mode will ensure that most
bustion. The subject matter is divided into seven main sections:
new utility power generation plants will also utilize the gas
turbine as the prime mover.
• An introduction to the main components of a conventional
When faced with new and challenging problems, the combus-
gas turbine combustor and their principal functions, along
tion engineer has traditionally looked to the fuel injector to
with brief descriptions of the types of fuel injectors now
provide a solution. This is not merely a pious hope, because in
in widespread use.
the past more than one problem that appeared at first sight to
• The properties of fuel sprays and recent developments in
constitute a fundamental barrier to further progress in combus-
the measurement and modeling of spray characteristics.
• A general overview of emissions concerns, their mecha-
Contributedby the InternationalGas TurbineInstituteand presentedat the 40th nisms of formation, and the methods employed to alleviate
InternationalGas Turbineand AeroengineCongress and Exhibition,Houston, pollutant emissions in conventional combustors.
Texas, June 5-8, 1995. Manuscriptreceived by the InternationalGas Turbine
InstituteApril 12, 1995.PaperNo. 95-GT-465.AssociateTechnicalEditor:C. J. • The use of variable geometry and staged combustion for
Russo. emissions reduction by control of flame temperature.

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 617
Copyright © 1995 by ASME
Downloaded From: on 01/27/2016 Terms of Use:
• Basic approaches to the design of " d r y " low NOx and ,,PRIMAR.y.=L__INTERMEDIATE_~.L__ DILUTION____,=
ultralow NOx combustors, with emphasis on the need to FUEL ZONE
feed the combustion zone with completely homogeneous
fuel-air mixtures, with the attendant problems of autoig- N
nition and flashback.
• Alternative methods for achieving ultralow NOx emis-
sions, notably rich-burn, quick-quench, lean-burn, and
catalytic combustors, both of which combine high poten-
tial for very low levels of nitric oxides with formidable
mixing and mechanical design problems.
• Correlation and modeling of nitric oxides and carbon
monoxide emissions.

Throughout the paper emphasis is placed on the important

and sometimes pivotal role played by the fuel preparation pro-
cess in all efforts to reduce pollutant emissions from gas tur- Fig. 1 Main components of a conventional gas turbine combustor
bines. Today the pollutant emission of primary concern is oxides
of nitrogen (NOx), and it is of interest to recall that combustors
upstream of the dilution zone. The role of the dilution zone is
designed specifically for low NOx first appeared on the scene
to admit the air remaining after the combustion and wall-cooling
in the 1970s, but remained "on the shelf" for another decade
requirements have been met, and to provide an outlet gas stream
due to a greater interest during that time in reducing the other
with a mean temperature and a temperature distribution that are
important emissions of carbon monoxide (CO), unburned hy-
acceptable to the turbine.
drocarbons (UHC), and smoke.
The locations of the three main zones described above in
An attempt is made to address both aircraft and stationary
relation to the various combustor components and the air admis-
gas turbines because neither one is more or less important than
sion holes, are shown in Fig, 1.
the other in regard to their contribution to atmospheric pollution.
However, there are two compelling reasons why aircraft engines Liquid Fuel Injection. The liquid fuels employed in air-
should receive the most attention. One is that aircraft engines craft engines must first be atomized before being injected into
employ liquid fuels, which must first be atomized and vaporized the combustion zone. Fortunately, atomization is easy to accom-
before they can burn. These additional processes increase the plish; for most liquids, all that is needed is a high relative
degree of difficulty in achieving low emissions by almost an velocity between the liquid to be atomized and the surrounding
order of magnitude. The second reason is that natural g a s - - air or gas. Pressure atomizers accomplish this by discharging
the favored fuel for stationary gas turbines--has a peak flame the liquid at high velocity into a relatively slow-moving stream
temperature around 140°C less than that of the distillate fuels of air. An alternative approach is to expose a slow-moving
employed in aircraft engines. As the formation of NO~ is expo- liquid to a high-velocity airstream. The latter method is gener-
nentially dependent on temperature, this lower flame tempera- ally known as airblast atomization. Pressure atomizers are usu-
ture reduces NOx levels by a factor of around two. For these ally of the "dual-orifice" type, as shown in Fig. 2. As its
reasons the emissions goals set for stationary engines tend to name suggests, this atomizer incorporates two orifices, arranged
be much more severe than for aero-engines. concentrically. The inner primary orifice is small, while the
outer secondary orifice is of larger area. At low fuel flows, only
the primary operates and atomization quality is good because
Conventional Combustors injection pressures are fairly high. With increasing fuel flow,
There are two basic types of gas turbine combustor: tubular, the injection pressure rises and would soon become excessive
or can, and annular. A tubular chamber is composed of a cylin- but, at a predetermined pressure, a valve opens and fuel is
drical liner mounted concentrically inside a cylindrical casing. then conveyed to the secondary orifice. This arrangement gives
Most of the early jet engines featured tubular chambers, usually satisfactory atomization over a wide range of fuel flows without
in numbers varying from seven to sixteen per engine. For air- calling for excessive fuel delivery pressures. The principal ad-
craft applications, the tubular system is now regarded as too vantage of dual-orifice nozzles is good mechanical reliability.
long and too heavy and the annular system is now the automatic Their drawbacks include potential plugging of the small pas-
choice for all new engines. However, tubular chambers do have sages and orifices by contaminants in the fuel and an inate
significant advantages in terms of easy accessibility and mainte- tendency toward high concentrations of exhaust smoke at high
nance, and they are still preferred for large stationary engines combustion pressures.
in the 80-plus megawatt range, such as the General Electric The most widely used type of airblast atomizer is illustrated
MS7001 engine, which has ten separate cans. in Fig. 3. In this design the bulk fuel is first spread into a
The annular combustor comprises an annular liner mounted thin continuous sheet, a process called "prefilming," and then
concentrically inside an annular casing. In many ways it repre- exposed to high-velocity, swirling air streams on both sides of
sents an ideal form of chamber, since its "clean" aerodynamic the sheet. Prefilming airblast atomizers are capable of producing
layout results in a compact unit of lower pressure loss than other fine atomization and a virtually smoke-free exhaust at high
combustor designs. All modern aero-engines employ annular combustion pressures. The main drawbacks of airblast atomiz-
combustors, which means that modern aero-derived industrial ers are poor lean blowout performance and poor atomization
engines, such as the General Electric LM6000 and the Rolls dunng engine startup when air velocities are low. These prob-
Royce RB211, all have annular combustors. lems can be readily overcome by incorporating a pilot or pri-
The volume contained within the liner of a conventional com- mary nozzle. At low fuel flows, all of the fuel is supplied
bustor may be divided conveniently into three main zones: pri- through the pilot nozzle, and a well-atomized spray is obtained,
mary, intermediate, and dilution. The function of the primary giving efficient combustion at start-up and idling. At higher
zone is to anchor the flame and to provide sufficient time, tem- power settings, fuel is supplied to both the airblast and pilot
perature, and turbulence to achieve essentially complete com- nozzles. The relative amounts are such that at the highest fuel
bustion of the fuel. The hot combustion products issuing from flow conditions most of the fuel is supplied to the airblast atom-
the primary zone flow into the intermediate zone, which pro- izer. By this means, the performance requirements of wide burn-
vides a region in which combustion can proceed to completion ing range and low exhaust are both realized.

618 / Vol. 117, OCTOBER 1995 Transactions of the ASME

Downloaded From: on 01/27/2016 Terms of Use:

~ InlRliatlonTubll
1.6 ~ SMD = 50 Ira1,q = 3
. _ , _ . .

¢3 1.2

0.8 / ~ 0,q - 2
~ SllcOndl~NoIzle
Fig. 2 Dual-orifice, pressure-swirl atomizer (courtesy of the Parker
Hannifin Corporation)
00 I I ~ I ~l I
50 100 150 200 250
Gas Injection. For almost all stationary gas turbines the
fuel employed is natural gas. Some engines are fitted with dual- Fig.4 Influenceof dropsizedistributionparameteronfrequencydistri-
fuel nozzles, which allow both liquid and gaseous fuels to be butioncurve
used, but usually the liquid fuel is for backup purposes only.
The methods used to inject gaseous fuels tend to be simple as
there are no requirements for atomization and fuel evaporation. rived from basic principles. In consequence, the majority of
They include injecting the gas through plain orifices, slots, investigations into the drop sizes produced in atomization have
swirlers, and venturi nozzles. resulted in empirical equations for mean drop size. Lefebvre
The main problem with gas injection is that of achieving (1989) has reviewed the available equations for the mean drop
the optimal level of mixing between air, fuel, and combustion sizes produced by the types of atomizers employed in gas tur-
products in the combustion zone. Too high a mixing rate pro- bine combustion. They show that mean drop sizes are largely
duces narrow burning limits. On the other hand, if the rate of dependent on atomizer size, design features, and operating con-
mixing is too low, the system may be prone to low-frequency, ditions, and also on the physical properties of the liquids em-
combustion-induced pressure oscillations. On engines designed ployed and the density of the surrounding gaseous medium.
to operate on both gaseous and liquid fuels, it is important that Most of the equations for mean drop size published before
the gas flow pattern be matched to that of the liquid fuel; other- around 1970 should be regarded as suspect due to deficiencies
wise some variation in the temperature distribution of the com- in the methods available for drop size measurements. Even
bustor outlet gases would occur during changeover from one expressions based on accurate experimental data should only
fuel to the other. be used within the ranges of air properties, liquid properties,
and atomizer operating conditions employed in their derivation.
Fuel Spray Characteristics Extrapolation to other conditions is fraught with risk because
changes in any of these variables could produce a change in
The spray properties of most relevance to the formation of the mode of atomization, which could have a significant effect
pollutant emissions are mean drop size, drop size distribution, on the manner and extent to which variations in the relevant
cone angle, and patternation. Other properties of special impor- flow parameters affect the drop size distribution in the spray.
tance for the successful modeling of spray characteristics in-
clude droplet and gas velocities, droplet trajectories, and mass Drop Size Distribution. Owing to the random and chaotic
flux distributions. nature of the atomization process, the threads and ligaments
formed by the various mechanisms of jet and sheet disintegra-
Mean Drop Size. Various definitions of mean drop size tion vary widely in diameter, and their subsequent breakup
are available, of which the most widely used is the Sauter mean yields a correspondingly wide range of drop sizes. Most practi-
diameter (SMD), which represents the volume/surface ratio of cal atomizers produce droplets in the size range from a few
the liquid in the spray. Unfortunately, the physical processes microns up to several hundred microns.
involved in atomization are not yet sufficiently well understood As no complete theory has yet been developed to describe
for mean diameters to be expressed in terms of equations de- the hydrodynamic and aerodynamic processes involved when
jet and sheet disintegration occurs under normal atomizing con-
ditions, a number of functions have been proposed, based on
either probability or purely empirical considerations, that allow
the mathematical representation of measured drop size distribu-
tions. The simplest expression for drop size distribution is that
of Rosin and Rammler (1933). It may be expressed in the form
~ F shr°ud
1 - Q = exp - ( D / X ) q (1)
uterAit Swider
where Q is the fraction of the total volume contained in drops
Inner Ai uel Swider of diameter less than D, and X and q are constants. Thus, the
Rosin-Rammler relationship describes the drop size distribu-
tion in terms of the two parameters X and q. The exponent q
provides a measure of the spread of drop sizes. The higher the
value of q, the more uniform is the spray.
Figure 4 shows frequency distribution plots for two sprays,
both having the same SMD of 50 /zm, but in one case the
Fig.3 Prefllming airblast atomizer (courtesy of the Parker Hannifin Cor- distribution parameter q has a value of 2 and in the other case
poration) a value of 3. This figure clearly demonstrates that knowledge

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 619

Downloaded From: on 01/27/2016 Terms of Use:

of SMD alone is not sufficient to determine the spray evapora-
tion time. One spray contains no drops larger than 130 #m ,~_PL,MPa (psi) SIMPLEXATOMIZER
whereas the other contains drops up to 200 /zm in diameter. 0 1.03 (150) d o - 0.457 mm
This represents a difference in total spray evaporation time A 1.38 (200)
of more than two to one. Differences of this magnitude must 0.36 I"1 1.72 (250)
necessarily have a pronounced effect on the formation rates of
all the main pollutants.
Cone Angle. A major difficulty in the definition and mea-
surement of cone angle is that the spray cone has curved bound-
aries, owing to the effects of air interaction with the spray. To
overcome this problem, the cone angle is often given as the ~c
initial angle as measured close to the nozzle where the spray 0.18
contour is still in the form of a perfect cone. Rizk and Lefebvre
(1987) used their own results and those of other workers to
derive the following expression in which the cone angle, as 0.12
measured close to the nozzle, is expressed as the product of
two dimensionless groups:
0.06 -
2 0 = o, K - O l "S , A~ p L d oz P L,/ # L2,011
) " (2)
where K is the dimensionless geometric group A J D s d o . A~, is 00 1.0 2.0 3.0 4,0 5.0
the inlet port area, Ds is the swirl chamber diameter, do is the
diameter of the discharge orifice, AP L is the liquid injection /0/d0
pressure, and PL and #r are the liquid density and absolute Fig. 5 Influence of discharge orifice length/diameter ratio on circumfer-
viscosity, respectively. ential fuel distribution
This equation shows that spray cone angles widen with in-
crease in fuel injection pressure, corresponding to an increase
in fuel flow rate. This effect has been confirmed in a number situation that is highly detrimental to low-emissions combus-
of experimental studies; for example, Chen et al. (1992). Equa- tion. The term adopted by the gas turbine industry for the pur-
tion (2) also suggests that ambient air or gas pressure has no pose of defining spray distribution is "spray patternation" and
effect on cone angle, but this is only true in the region near the the instruments used to measure liquid flux distributions in
nozzle. As the droplets move away from the nozzle they are sprays are commonly referred to as "spray patternators."
subjected to radial air currents generated by their own kinetic The simplest form of patternator comprises a cluster of col-
energy, as described by De Corso and Kemeny (1957). These lection tubes, which are located under the test nozzle. The noz-
air currents become stronger with increase in ambient air pres- zle sprays downward into the tubes and the duration of each
sure, with the result that although the cone angle as measured test is the time required for one of the sampling tubes to become
close to the nozzle does not change appreciably as the pressure nearly full. After the level of fuel in each tube is measured and
is increased, the spray width further downstream contracts recorded, the values are averaged to get a mean height. The
markedly. This decrease in spray cone angle with increase in levels of the tubes are normalized against the mean, and the
ambient pressure is one of the main reasons why combustors standard deviation of the normalized values is calculated. The
featuring pressure-swirl atomizers generate more soot and normalized standard deviation provides a quantitative indication
smoke as the combustion pressure is increased. of the symmetry of the distribution of fuel within the spray.
Airblast atomizers are much less susceptible to variations in More sophisticated spray patternators have been devised that
fuel and ambient air pressure than pressure-swirl atomizers. are capable of high-resolution measurements of the mass flux
The liquid sheet or jet exposed to the atomizing air has little distributions produced by gas turbine fuel injectors (McVey et
momentum and the droplets formed in atomization are entirely al., 1987, 1989). These methods are not necessarily more accu-
dependent on the kinetic energy of the atomizing air to transport rate than the simple patternators described above, but they do
them away from the nozzle. This means that droplet trajectories allow a large amount of data to be collected in a relatively short
are dictated by the air movements created by air swirlers and time.
other aerodynamic devices, which form an integral part of the Circumferential patternation is sensitive to the dimensional
nozzle configuration. Thus, airblast atomizers are characterized relationships characteristic of a particular nozzle design. For
by a lack of sensitivity of spray geometry characteristics to the example, for pressure-swirl atomizers, patternation has been
physical properties of both the liquid and the surrounding gas- correlated with the degree of eccentricity between the swirl
eous medium. chamber and the final discharge orifice (Tate, 1960). Nozzle
From tests carried out on prefilming airblast atomizers, Ort- quality is also important, and spray patternation may be im-
man and Lefebvre (1985) found that increasing the air pressure paired by poor surface finish, orifice imperfections, plugged or
differential across the nozzle caused the spray angle to contract contaminated flow passages, eccentric alignment of key nozzle
slightly. Custer and Rizk (1988) observed a similar effect. For components, and other conditions. For alrblast atomizers, lack
one prefilming atomizer, operating at an air/fuel ratio of 5, they of symmetry in the various air passages and swirlers can also
found that a large increase in air pressure differential from 1 have an adverse effect on spray patternation, as discussed below.
to 4 percent caused the spray angle to be reduced from 100 to Chert et al. (1993) used several different pressure-swirl noz-
80 deg. zles to examine the effects of variations in liquid properties,
operating conditions, and atomizer design features, on spray
Circumferential Fuel Distribution, Drop sizes are, of patternation. Of special interest are the results obtained on the
course, of great importance in low-emission combustion be- effects of varying the length/diameter ratio of the final dis-
cause of their strong influence on fuel evaporation rates and charge orifice, lo/do, as illustrated in Fig. 5. In this figure the
droplet lifetimes. Of equal importance, however, is the symme- circumferential maldistribution is expressed in terms of a stan-
try of the spatial distribution of fuel within the spray. Failure dard deviation, a. If the circumferential distribution of liquid
to achieve a symmetric mass flux distribution can result in local within the spray were completely uniform, the value of a would
regions of mixture inhomogeneity in the combustion zone, a be zero. Figure 5 shows that a declines, i.e., patternation ira-

620 / Vol. 117, OCTOBER 1995 Transactions of the ASME

Downloaded From: on 01/27/2016 Terms of Use:

proves, with increase in liquid injection pressure, presumably CONFIGURATION
because high injection pressures promote more turbulence and
better mixing in the swirl chamber. It also shows that the cir-
cumferential distribution is most uniform for a value of lo/do CONCENTRIC
of 2. This optimum value of 2 was found to apply at all operating
conditions to all of the liquids tested, some of which varied in
viscosity by a factor of twelve. This result is of considerable
practical interest because for many years the trend in pressure-

J i ",,
swirl atomizer design has been toward lower values of It~do in
order to reduce internal losses and thereby improve atomization.
Most current atomizer designs have values of lo/do of 0.5 or
Rp= 9.7mm A
less. Figure 5 clearly demonstrates that reducing It~do can appre-
ciably worsen the spray patternation.
Spray Diagnostics. In recent years considerable advances RADIALSWlRLER
have been made in the development of laser diagnostic tech- FLARE
niques for measuring the properties of sprays. The methods
used in spray analysis have been reviewed by Chigier (1983),
Chigier and Stewart (1984), and Durst et al. (1981). Laser-
Doppler velocimetry and phase-Doppler anemometry, inter-
faced with signal processing and high-speed data acquisition
systems, are now being used routinely in simultaneous measure- . .,.ARYsw,RL
ments of drop size, drop velocity, and local drop drag coeffi- + +++.!"
..,I=" ®
cients (Bauckhage, 1988; Delarosa et al., 1993; Dittmann et al.,
1992; Naqwi et al., 1991). High-speed pulsed microphotogra- I~ ....... 'R- ,/~1~/~ SECONDARY
phy, cinematography, and holography are also used to study
drop size distributions and spray structure (Santangello and
Sojka, 1994). Much of the tedium normally involved in detailed
tz+ "
studies of drop size distributions in various regions of the spray • NOZZLE
can now be alleviated using automatic image analysis, in which VENTURI
droplet images are enlarged, counted, and sorted by electronic (3)RADIALSWlRLER
scanning devices and analyzed by microprocessors. 0.125"
Many efforts are now underway to gain a better understanding
of the basic phenomena occurring within complex practical sys- Fig. 6 Concentric and eccentric airblast atomizers
tems currently in service on advanced turbojet engines. Wang
et al. (1992a, b, 1994, 1995) and McDonell et al. (1994a, b)
used a two-component phase-Doppler interferometry system to downstream. The slight flow asymmetries observed with con-
make detailed measurements in the flows downstream of a centric alignment of the swirlers is attributed to normal hard-
SNECMA/GE CFM-56 airblast atomizer. In addition to time- ware tolerance. The measurements obtained with the eccentric
averaged properties, transient phenomena were also examined. swirler assembly exhibit asymmetries in both mean velocities
These studies have provided great insight into the general struc- and turbulence.
ture of the gas phase as well as details regarding the complex Hebrard et al. (1991, 1993) also used two-component phase-
behavior of drops. Of special interest from a practical viewpoint Doppler interferometry to characterize the two-phase flow field
is that actual engine hardware was used to examine the sensitiv- generated by an engine airblast atomizer. Their measurements
ity of the spray structure to scaling and hardware variations. included mean axial and transverse velocities for each phase as
One study was devoted to an examination of the effects of well as volume flux and drop size distributions. They used
nozzle defects on spray characteristics. Figure 6 presents a sche- surface plots to illustrate the local variation of Sauter mean
matic diagram of the CFM-56 atomizer and swirl cup used in diameter for cross sections of the spray at several different
this investigation. It features a pair of concentric air inlets, distances downstream from the atomizer. Close to the nozzle,
which generate two separate swirling airflows in opposite direc- they observed a dome-shaped distribution of drop sizes. Air
tions. The swirl cup acts as a "hybrid" injector, with drops flowing through the inner and outer swirlers of the airblast
produced from a combination of airblast and pressure atomiza- atomizer produced a nearly uniform drop size distribution, but
tion. This figure illustrates what can happen to the swirl cup recirculating air jets caused some secondary atomization. Some
assembly during the engine operating cycle. The design calls large drops were observed in the core, generated by the inner
for concentric alignment at altitude cruise conditions. However, swirling air. Farther downstream the drop size distribution
due to "throttle pushes" over the years, the assembly can be- broadened and indentations appeared on the spray boundary,
come eccentric by up to around 3 mm (0.125 in.), as indicated caused by the presence of air jets emanating from the holes
in Fig. 6. This occurs because the secondary swirler is attached located near the edge of the atomizer cup.
to the combustor liner, whereas the burner feed arm is mounted Wynne and Jasuja (1991) and Jasuja and Lefebvre (1994)
directly onto the combustor casing. Since the two components used various nonintrusive techniques, including phase-Doppler
are pinned at two different locations to the casing, the thermal interferometry, high-intensity spark photography, high-speed
expansion of the casing causes misalignment. cinephotography, and video imaging to capture the dynamic and
Results showing the effect of misalignment on the flow field unsteady spray characteristics produced in prefilming airblast
downstream of the atomizer are presented in Fig. 7. In this atomization. All of the injectors used in this program were
figure the two circles on each cartoon represent the primary typical of modern gas turbine practice and differed from each
and secondary swirlers. The vector quantities represent the gas other only in regard to various design features, such as swirler
velocities in the r - Z plane, while the contours are the turbulent flow area and vane angle, m special feature of this work is that
kinetic energies based on the measured rms velocity compo- it covers a very wide range of ambient air pressures. The results
nents. The highest velocity variations occur in the shear layer obtained are generally consistent with the findings reported
between the recirculating flow and the surrounding flow moving above in showing that the sprays produced by actual engine

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 621

Downloaded From: on 01/27/2016 Terms of Use:

GAS PHASE STRUCTURE grid to compute the internal flow characteristics of an injector.
Their calculations were compared with LDV measurements on
four nozzle configurations with variations in size, swirl strength,
Klneffo End and swirl direction. For most mean velocity components, the
1311 agreement between measurements and calculations was good.
aSt These workers concluded that the modeling results indicate that
....... 34t
~ 0 to calculations can be used for design and analysis.
Hebrard et al. (1993) have reported on the development of
a similar k - c code, which shows good correlations between
E Z = 1.7S Rp
predictions and measurements of drop trajectories from an air-
blast atomizer. Data were taken using a high-speed video cam-
Z - 2.75 Rp era. A Weber number relationship was derived for predicting
Q Z = 3.7S Rp whether a droplet will bounce or film when hitting a wall, and
it was also noted that fuel droplets are capable of burning either
Z = 4.7S Rp singly, in groups, or en masse.
•50 -40 -30 -20 -10 O 10 20 30 40 50 The examples given above are intended to illustrate current
RADIAL POSITION, mm efforts in fuel injector modeling. These models, in conjunction
with the detailed knowledge on spray characteristics that new
"§ I 20 ~i'L advances in laser diagnostics are making available at an ever-
~ ~'" ~.~:'~ !: :i:(~ Z = 1.7S Rp increasing rate, are providing better insight on the critical role
of the fuel injection process on low-emissions combustor perfor-

I Fuel Nozzle Coking

.-50 -40 -30 -20 -10 0 10 20 30 40 SO
RADIAL P o s m o N , mm
No discussion on spray characteristics would be complete
without reference to the deleterious effects on atomization and
Fig. 7 Influence of atomizer misalignment on flow field spray symmetry caused by the partial or complete blockage by
carbon deposition of one or more of the flow passages within
the fuel nozzle. This problem is one of growing importance and
nozzles exhibit appreciable nonuniformities in regard to droplet has special implications for the development of many types of
trajectories and mass flux distributions. low-emission combustors for which a uniform distribution of
A number of laboratories throughout the world are now using fuel within the spray is a basic requirement. It is especially
advanced laser diagnostics to characterize the sprays produced serious for advanced turbojet engines due to the growing use
in airblast atomization. Much of this work is confined to dilute of the fuel as a heat sink for cooling the airframe, avionics, and
and simple sprays operating in easily probed laboratory condi- engine lubricating oil. The problem is further exacerbated by
tions at normal atmospheric pressure. The special merit of the the fact that the fuel feed arm is immersed in the compressor
researches summarized above is their focus on full-scale engine efflux air. This high-temperature, high-pressure, high-velocity
nozzles, operating in some cases over wide ranges of ambient airflow causes convective heating, which further raises the tem-
air pressure. The information they are providing on the complex, perature of the fuel before it flows into the fuel injector. The
near-field bulk liquid behavior and droplet physical processes combined effect of all these various inputs is that by the time
is making a significant contribution to fuel injector modeling the fuel is sprayed into the combustion zone, its temperature is
capabilities and the development of low-emissions combustors. appreciably higher than when it left the fuel tank.
From a combustion viewpoint this elevation in fuel tempera-
Modeling of Spray Characteristics. The development of ture is not altogether undesirable because it reduces fuel viscos-
new diagnostic techniques as discussed above has encouraged ity and thereby promotes finer atomization. Unfortunately, high
the formulation of empirical and analytical models to describe fuel temperatures stimulate oxidation reactions, which lead to
the fuel injection process. Rizk and his co-workers (see, for the formation of gums and other insoluble materials (including
example, Rizk et al., 1987; Custer and Rizk, 1988; and Rizk carbon), which tend to deposit on the walls of the passages and
and Mongia, 1992a, b) have been especially active in this field. metering orifices within the nozzle. These deposits can distort
One overall model, which combines three submodels, addresses the fuel spray and create appreciable nonuniformities in spray
various processes of fuel and air preparation for atomization pattemation.
and formation of the final spray. The first submodel describes The problems created by the deposition of carbonaceous ma-
the velocity profiles of the air passing through and exiting from terials, generally referred to as "coke," within the fuel nozzle
the nozzle passages. The second submodel deals with the fuel are of special importance for small gas turbines. As the engine
prefilming process and the conversion of the thin fuel sheet size is reduced, the dimensions of various engine components
into droplets. Finally, the third model represents the turbulent must also decrease. This is particularly disadvantageous for the
dispersion effects on the spray. fuel injector because it embodies small internal flow passages,
The results of this model highlight the need to define both which are especially prone to plugging and blockage. In small
the air flowfield and the film breakup process with good accu- gas turbines, the swirl-inducing and fuel metering passages can
racy. The model also indicates how improvements in atomiza- have linear dimensions as small as 200/~m (McCaldon et al.,
tion quality can be achieved by careful design of the prefilmer 1993). Fuel passages are often convoluted, thus creating local
and internal passages of the atomizer. A thinner fuel film, which, regions in which the flow velocity is low and thermal degrada-
for any given fuel flow rate, is governed by local air velocity tion and coking can occur. The pressure-swirl atomizer, having
and nozzle geometry, results in shorter ligaments that disinte- very small hole and passage sizes, is especially prone to coke
grate into drops closer to the nozzle in regions of relatively formation. Coke agglomerates, formed either upstream of the
high air velocity. This enhances the secondary atomization pro- nozzle tip or within the nozzle itself, can break off and be
cess and reduces the final drop sizes in the spray. carried in to the metering passages.
Sultanian and Mongia (1986) used the k - e turbulence model Airblast atomizers are inherently less susceptible than pres-
in conjunction with a boundary-fitted curvilinear orthogonal sure-swirl atomizers to the problems of fuel coking because

622 / Vol. 117, OCTOBER 1995 Transactions of the A S M E

Downloaded From: on 01/27/2016 Terms of Use:

they employ much larger fuel passages in the nozzle tip. How- Table 1 Principalpollutantsemitted by gas turbines
ever, the inability of airblast atomizers always to meet the re-
quirements of cold day starting has prevented pure airblast sys- POLLUTANT EFFECT
tems from completely displacing pressure-swirl atomizers from
engine designs.
The effects of partially or totally blocked fuel metering pas- Carbon monoxide (CO) Toxic
sages on the fuel-air distributions produced by an airblast atom-
izer have been examined by McCaldon et al. (1993). They Unburned hydrocarbons (UHC) Contributes to urban smog
found that as more and more fuel metering holes become ob-
structed with increasing operating time, more fuel is forced
Particulate ma~er (C) Visible. Possible link to
through the remaining nozzles. Consequently, engine damage
may be caused by those injectors, which, if tested individually, asthma and other respiratory
still flow within tolerances. Although the report of McCaldon diseases
et al. contains no data on NOx, it is clear that the fuel flux
variations found must also give rise to substantial increases in Oxides of nitrogen (NOx) Toxic, precursor of chemical
NOx emissions at high power conditions and increased emis- smog, depletion of ozone in
sions of CO and UHC at low power conditions. stratosphere
The studies carried out by Nickolaus and Lefebvre (1987)
and McCaldon et al. (1993), among others, have demonstrated Oxides of sulfur (Sex) Toxic, corrosive
that coke formation within fuel nozzles can have a serious ad-
verse effect on spray pattern. Although it has long been recog- Carbon dioxide (CO2) Contributes to greenhouse
nized that "cleaner" fuels of low aromatics content can greatly
reduce the potential for fuel coking, this has never been regarded effect
as a practical option due to the requirements of fuel flexibility
for the military and lower-cost, heavier distillate fuels for civil-
ian and commercial applications. passage flow area, reduced surface roughness, additional insu-
The discussion above on the effects of fuel thermal degrada- lating air gaps, and replacement of metallic tip components with
tion on fuel injector performance has tended to focus on the ceramics minimized the wetted-wall temperature. Vacuum gaps,
deleterious effects caused by the elevation in temperature expe- reduced emissivity coatings, and thermal barrier coatings did
rienced by the fuel upstream of the nozzle tip. Another important not offer significant wall-temperature reductions.
effect on fuel coking is the fuel-wetted wall temperature within Thermal modeling studies carried out by Myers et al. (1992)
the nozzle. An increase in wall temperature greatly accelerates suggested that the two major sources of heat absorption into
the rate of coke deposition. The flow of heat into the nozzle tip the fuel nozzle were the air swirl vanes and any surface exposed
by air flows introduced to prevent soot accumulation, and by to the flame. At an altitude cruise condition, for example, the
intense radiation from the combustion gases, cause fuel-wetted predicted heat flux entering the nozzle face from flame radiation
wall temperatures to be appreciably higher than the temperature is more than 20 times that absorbed by conduction and convec-
of the adjacent fuel. This temperature gradient near the wall not tion through the burner feed arm. The frontal area exposed to
only enhances deposition rates by raising the fuel temperature the flame is thus a key element in nozzle thermal loading.
but also has a separate additional effect due to the higher wall Myers et al. concluded that substantial reductions in wetted-
temperature. wall temperatures can be realized at extreme fuel and air inlet
In recent years the U.S. Naval Air Propulsion Center has temperatures by using simple air gaps as thermal barriers. De-
sponsored an innovative high-temperature fuel nozzle program tailed thermal analysis and simple thermal barriers, rather than
with the objective of designing and evaluating fuel nozzles exotic cooling schemes, can produce dramatic improvements in
capable of operating satisfactorily despite extreme fuel and air thermal protection.
inlet temperatures. As part of this program, Stickles et al. ( 1993 ) The problem of fuel coking and its strong adverse effects on
evaluated 27 different nozzle designs, all of which were based spray uniformity and pollutant emissions is one of growing
on the production GE F404 fuel nozzle. The design approach concern due to the anticipated gradual deteriorations in fuel
was to use improved thermal protection and fuel passage geom- quality and the continuing trends toward higher temperature
etry in combination with fuel passage surface treatment to mini- engines.
mize coking. Heat transfer analysis highlighted several im-
portant design features for reducing wetted-wall temperatures.
Emissions Concerns
1 Reduce fuel flow passage area to increase fuel velocity The exhaust from an aircraft gas turbine is composed of
* Increases heat transfer coefficients carbon monoxide (CO), carbon dioxide (CO2), water vapor
* Reduces wetted-wall temperature (H20), unburned hydrocarbons (UHC), particulate matter
* Reduces fuel residence time (mainly carbon), oxides of nitrogen, and excess atmospheric
2 Add air gaps oxygen and nitrogen. Carbon dioxide and water vapor have not
* Metals with air gaps are more effective than solid always been regarded as pollutants because they are the natural
ceramics consequence of complete combustion of a hydrocarbon fuel.
3 Substitute ceramics for metal parts However, they both contribute to global warming and can only
* Ceramics are very effective in combination with air be reduced by burning less fuel. With industrial gas turbines
gaps burning residual fuels, an additional pollutant of concern is
4 Avoid bends and steps in the fuel flow path oxides of sulfur, mainly SO2 and SO3. They are toxic and corro-
* They create stagnant fuel regions that have low trans- sive and lead to the formation of sulfuric acid in the atmosphere.
fer coefficients and thus high wetted-wall tempera- Since virtually all the sulfur in the fuel is oxidized to SOx, the
tures only viable limitation strategy is to remove sulfur from the fuel
Sample tube coking test results showed the importance of prior to combustion.
surface finish on the fuel coklng rate. Reducing the surface The principal pollutants are listed in Table 1. Carbon monox-
roughness from 3.1 #m to 0.25 #m reduced the deposition rate ide reduces the capacity of the blood to absorb oxygen and, in
by 26 percent. In summary, Stickles et al. found that reduced high concentrations, can cause asphyxiation and even death.

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 623

Downloaded From: on 01/27/2016 Terms of Use:

Unburned hydrocarbons are not only toxic, but they also com- Table 3 Typical aircraft emissions
bine with oxides of nitrogen to form photochemical smog. Par-
ticulate matter, often called soot or smoke, is generally not
considered toxic at the levels emitted, but recent studies by TYPICAL DISTRIBUTIONOF TOTAL EMISSION MASS
Seaton et al. (1995) indicate a strong association between AIRCRAFTEQUIPPED WITH MODERN ENGINES
asthma and other respiratory diseases and atmospheric pollution
by concentrations of small particles in the microgram range.
Oxides of nitrogen (NO + NO2), of which the predominant o Aircraft: Twin-engine transport
o Range: 500 nautical miles
compound at high emission levels is NO, not 0nly contribute
to the production of photochemical smog at ground level but
also cause damage to plant life and add to the problem of acid Category Percent of total emission mass
rain. Relative to other sources, aircraft engines are only minor
contributors to the overall NOx burden. For example, in the , During ICAO ! During elimbout/ Overall
United States NO~ emissions from aircraft engines account for landing-takeoff cruise/descent
only about 0.5 percent of the total emissions nationwide from cycle
all sources (Bahr, 1991). On a global basis, NOx emissions
from aircraft engines constitute less than 3 percent of all man- Smoke 0.1 0.1
made NOx emissions. However, of special concern is that these
emissions are released at altitude and lead to the formation of HC 0.6 t.0 1.6
ozone in the troposphere. This is the region in which stationary
CO 5.4 7.0 12.4
gas turbines and subsonic aircraft operate. Measurements taken
over a long period of time at altitudes from one to three kilome- NOx 7.8 78.1 85.9
ters indicate that the level of ozone over Western Europe is now
approaching 50 ppb (parts per billion). Prolonged exposure to
ozone concentrations around 100 ppb is associated with respira-
Total 13.8 86.2 I00.0
tory illnesses, impaired vision, headaches, and allergies. Ground
(56.5% NOx) (90.6%NOx)
level ozone is especially important in regions where the topo-
graphical features prevent the local weather system from remov-
ing the ozone formed in combustion and where strong sunshine
can promote the photochemical reactions that lead to smog. Los
Angeles is the classic example of such a region. It is hardly an operational cycle around airports. This LTO cycle is intended
surprising therefore that the drive toward very stringent emis- to be representative of operations performed by an aircraft as
sions legislation on NOx first emanated from this city. it descends from an altitude of 914 m (3000 ft) on its approach
Similar studies indicate that NOx emissions emitted at higher path to the time it subsequently attains the same altitude after
altitudes can contribute to the depletion of the stratospheric takeoff.
ozone layer. Unfortunately, supersonic aircraft are required to The standards for CO, UHC, and NOx emissions are presented
operate at such high altitudes. Ozone depletion results in an in Table 2, in which 7r0ois the engine pressure ratio at takeoff.
increase in ground level ultraviolet radiation, which leads to an They are expressed in terms of a parameter that consists of the
increase in the incidence of skin cancer. total mass in grams of any given gaseous pollutant emitted
For industrial gas turbines the problems posed by exhaust during the LTO cycle per kilonewton of rated thrust at sea level.
gas emissions are no less challenging than for aero-engines. We have
World energy demand is forecast to grow over the next 30 years Emission = Emission Index (EI)
at around 1.4 percent per annum. This demand will be met
predominantly by the combustion of fossil fuels (Singh, 1994). g/kN g/kg fuel
Thus the manufacturers and users of gas turbines for utility
power generation now find themselves at the forefront in regard × Engine SFC × Time in Mode
to responsibility for emissions issues. (3)
kg fuel/h kN h

Emissions Regulations This equation demonstrates that improving (i.e., reducing) the
specific fuel consumption results directly in a reduction of the
The International Civil Aviation Organization (ICAO) has pollution levels. Because the CO and UHC levels of modem
promulgated regulations for civil subsonic turbojet/turbofan en- engines have been greatly reduced at all low power conditions,
gines with rated thrust levels above 26.7 kN (6,000 pounds) and only NOx is emitted in appreciable amounts at altitude
for a defined landing-takeoff cycle (LTO) which is based on cruise, in practice the emissions generated by aircraft engines
consist primarily of NOx. A typical example of the emissions
Table 2 ICAO gaseous emissions standards mass distribution associated with the flight of a modem subsonic
aircraft has been provided by Bahr (1992) and is shown in
Emission, Subsonic turbojet/ Supersonic turbojet/ Table 3. This table represents a flight of 900 km (500 nautical
g/kN turbofan engines* turbofan engines miles) and shows that NOx emissions predominate both in the
vicinity of the airport and during altitude cruise. For a longer
flight, NOx emissions would account for an even larger fraction
o I-K2 19.6 140(0.92)~r,0o of the total emissions mass.
o CO 118.0 4550(~oo)-I.03 The ICAO standard for smoke measurement is expressed in
terms of a Smoke Number (SN), which is related to the engine
o NOx 40+2goo 36+2.421r.oo takeoff thrust (F00) by the expression
*Newly manufactured engines with rated takeoff SN = 83.6(F0o) -°'274 (4)
thrust greater than 26.7 kN.
This expression is shown graphically in Fig. 8. The intention
~oo = engine pressure ratio at takeoff of this standard is to eliminate any visible smoke from the
engine exhaust. As smoke visibility depends on both the smoke

6 2 4 / Vol. 117, O C T O B E R 1995 T r a n s a c t i o n s of the A S M E

Downloaded From: on 01/27/2016 Terms of Use:

/^ I NO,,


Ul 40

z 30 el = 83.6 (Fo0)-0.274
I I co
I"i ~ H C
0 50 100 150 200 250 300
RATED THRUST (F0o), kN Fig. 10 Emissions characteristics of gas turbine engines

Fig. 8 ICAO smoke emissions standards

Singh (1994), summarizes the EEC Directive of August 19,
concentration, as indicated by the value of SN, and on the 1992, concerning CO and NOx emissions for gaseous and liquid
viewing path length, the allowable SN of a high thrust engine fuels for gas turbines over and under 50 MW. These levels are
is lower than for a low thrust engine because of its larger exhaust then compared with some legislation from the U.S. and Japan.
diameter. Most of the drive toward more stringent regulations for sta-
As far as subsonic aircraft engines are concerned, the situa- tionary gas turbines has been directed at oxides of nitrogen.
tion in regard to compliance with ICAO regulations is generally Changes in legislation are now calling for NOx requirements as
satisfactory, due mainly to the engine manufacturers' efforts low as 25 ppm and even 9 ppm. In some locations the installa-
during the past 20 years in improving combustor design and in tion of a new gas turbine power generating facility calls for
reducing engine specific fuel consumption. However, in view NOx emissions levels below 5 ppm. Moreover, there is often a
of the continuing pressure to reduce NOx emissions from all requirement to use the "Best Available Control Technology"
sources, efforts by ICAO to formulate more rigorous NOx stan- (BACT) or the "Lowest Available Emission Rate" (LAER).
dards are in progress and a 20 percent increase in the stringency This has led to concerns such as those expressed by Angello
of the current ICAO standard has been enacted and will come and Lowe (1989), that if a new technology is developed that
into effect at the end of 1995. In Sweden a tax is now imposed significantly improves the ability to reduce NOx emissions, it
on NOx and UHC emissions generated during domestic flights, effectively sets the emission standard that all subsequent plants
while limits on the total yearly emissions quantities at some must meet. Thus, as new technologies are developed to meet
airports are being considered in both Sweden and Switzerland the ever-increasingly restrictive emission limits, they become
(Bahr, 1995). These steps could constitute important prece- the standard by which the next round of emission regulations
dents. For future supersonic aircraft engines, very stringent is guided.
goals have been established. The future of second-generation Combustors designed specifically for low-NOx emissions are
supersonic aircraft depends crucially on compliance with these often described as "low-NOx" or "ultralow-NOx". Broadly
goals, which can only be met by the use of yet-to-be-developed speaking, the term "low-NOx" relates to emissions in the range
ultralow NOx combustor designs. from 25 to 50 ppm (corrected to 15 percent O2), whereas "ul-
The discussion so far has focused mainly on aircraft engines, tralow NOx" generally denotes emissions below 25 ppm. With
but the same general considerations apply with equal force to the advances in low emissions technology, the description "ul-
use of gas turbines in other applications. Regulations governing tralow-NOx" is becoming increasingly reserved for emission
emissions from gas turbines at ground level tend to be highly levels below 10 ppm.
complex because the legislation varies from one country to
another and is supplemented by local or site-specific regulations Mechanisms of Pollutant Formation
and ordinances governing the size and usage of the plant under The concentration levels of pollutants in gas turbine exhausts
consideration and the type of fuel to be used. Figure 9, due to can be related directly to the temperature, time, and concentra-
tion histories of the combustion process. These vary fi'om one
combustor to another and, for any given combustor, with
SOME EMISSIONS REGULATIONS changes in operating conditions. The nature of pollutant forma-
tion is such that the concentrations of carbon monoxide and
• NOX unburned hydrocarbons are highest at low-power conditions and
7o diminish with increase in power. In contrast, oxides of nitrogen

. ~ 40
60 NOX

• CO
and smoke are fairly insignificant at low power settings and
attain maximum values at the highest power condition. These
characteristic trends are sketched in Fig. 10.
(LIQUID) Carbon Monoxide. In combustion zones designed to oper-
o> 2o ate fuel-rich, large amounts of CO are formed owing to the lack
• CO
10 (GAS) of sufficient oxygen to complete the reaction to CO2. If, how-
0 ever, the combustion zone mixture strength is stoichiometric or
>50 MW <50 MW US Japan
moderately fuel-lean, significant amounts of CO will be present
owing to the dissociation of CO2. In principle, it should be
Fig. 9 Some emissions regulations for industrial engines possible to reduce this CO to a very low level by the staged

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 625

Downloaded From: on 01/27/2016 Terms of Use:

However, at temperatures higher than around 1800 K the pro-
200 duction of CO by chemical dissociation of CO2 starts to become
significant. Thus, at high equivalence ratios CO levels are high
because they attain equilibrium values. Only in the fairly narrow
range of equivalence ratios from around 0.7 to 0.9 can low
100 levels of CO be achieved.
Measurements of CO emissions from more practical combus-
tion chambers are fully consistent with these results. For exam-
60 pie, Sturgess et al. (1993) used a JT9D-70A combustor supplied
with aviation fuel to examine the effects of combustor operating
conditions and airflow distribution on the emissions of CO,
,~ 40 UHC, NOx, and smoke. The results obtained for CO at engine
idle conditions are shown in Fig. 12. Although this figure dem-
onstrates minimum CO at an equivalence ratio around 1.0 as
opposed to the value of 0.8 obtained by Rink and Lefebvre,
20 FUEL - DF 2 this difference is not considered significant because the amount
S M D - 110 pm of air participating in primary combustion is notoriously diffi-
P^ - 1.27 MPa cult to estimate accurately.

Influence oflnlet Air Temperature. The effect of an increase
o T A = 473 K in inlet air temperature is to reduce CO, as illustrated in Fig.
8 E1 T A . 573 K 11. This it does by raising the flame temperature, which acceler-
ates the conversion of CO into CO2. At equivalence ratios of
6 stoichiometdc and above, this higher flame temperature pro-
0.6 0.8 1.0 1.2 1.4 motes the formation of CO by dissociation so that, at the highest
equivalence ratios, the CO concentrations are highest for the
EQUIVALENCE RATIO, higher inlet air temperature. These trends are clearly illustrated
in Fig. 11.
Fig. 11 Influence of inlet air temperature on CO emissions
Influence of Pressure. Figure 13 demonstrates the beneficial
effect of an increase in combustion pressure in reducing CO
admission of additional air downstream of the primary zone to emissions. At low equivalence ratios, increase in pressure di-
achieve a gradual reduction in burned gas temperature. In prac- minishes CO by accelerating the rate of conversion of CO into
tice, CO emissions are found to be much higher than predicted CO2. At high equivalence ratios, increase in combustion pres-
from equilibrium calculations and to be highest at low-power sure reduces CO emissions, albeit to a lesser extent, by sup-
conditions, where burning rates and peak temperatures are rela- pressing chemical dissociation.
tively low.
Influence of Ambient Air Temperature. Hung and Agan
The main factors influencing burning rates, and hence also
(1985) have examined the influence of ambient air temperature
CO emissions, are engine and combustor inlet temperatures,
on the CO emissions from a 7 MW industrial engine supplied
combustion pressure, and primary-zone equivalence ratio. (Note
with natural gas fuel. A strong air temperature effect on mea-
that equivalence ratio is the actual fuel/air ratio divided by the
sured CO was observed. CO emissions for an air temperature
stoichiometric value of fuel/air ratio.) All these aspects have
been investigated by many workers, including Bahr (1982,
1987, 1991, 1992), Bayle Labour6 (1991), Correa (1991),
Desaulty ( 1991 ), Koff (1993), Mosier and Pierce (1980), and 40
Sturgess et al. (1993). In their studies on pollutant emissions
from continuous flow combustors, Rink and Lefebvre (1989a) P&W JT9D-70A
used a continuous flow tubular combustor, 150 mm in diameter, 85 IDLE
in conjunction with an array of 36 equally spaced "micro-
scopic" airblast atomizers, to achieve a uniform distribution of
fuel in the mixture entering the combustion zone. This method 30 -
of fuel injection had another useful advantage in that it allowed
the mean drop size in the fuel spray to be varied in a controlled ._
manner while maintaining all other flow conditions constant. ~. 25 -
All measurements of pollutant emissions were carried out at a 9)
distance of 170 mm from the fuel injectors. 5,
Influence of Equivalence Ratio. Some of the results ob- ~ 20-
tained by Rink and Lefebvre are presented in Fig. 11, which
shows the variation of CO emissions with equivalence ratio for
two values of inlet air temperature, namely 473 and 573 K. 15 -
Both curves exhibit the same general characteristics. They show
that CO emissions diminish with increase in equivalence ratio,
reaching a minimum value at an equivalence ratio, ~b, of around 10 - O
0.8. Beyond this point any further increase in equivalence ratio
causes CO emissions to rise. These trends are typical of those
observed for other types of combustion systems. The high levels 5 = I I I I
of CO at low equivalence ratios are due to the slow rates of 0.4 0.6 0.8 1.0 1.2 1.4
oxidation associated with low combustion temperatures. In- PRIMARY ZONE EQUIVALENCE RATIO
crease in equivalence ratio raises the flame temperature, which
accelerates the rate of oxidation so that CO emissions decline. Fig. 12 Influence of primary zone equivalence ratio on co emissions

626 / Vol. "117, OCTOBER 1995 Transactions of the A S M E

Downloaded From: on 01/27/2016 Terms of Use:

in much the same manner. Thus, in common with CO, UHC
emissions are reduced by increases in inlet air temperature and
combustor pressure, and by any change in equivalence ratio
toward the optimum value of around 0.8.
Smoke. Exhaust smoke is caused by the production of
finely divided soot particles in fuel-rich regions of the flame.
With pressure-swirl atomizers, the main soot-forming region
80- lies inside the fuel spray, at the center of the combustor, as
illustrated in Fig. 14. This is the region in which the recirculat-
60- ing burned products move upstream toward the fuel spray, and
J' in which local pockets of fuel vapor are enveloped in oxygen-
40- deficient gases at high temperature. In these fuel-rich regions,
0 soot may be produced in considerable quantities. Most of the
d¢J soot produced in the primary zone is consumed in the high-
~...j~ / FUEL,OF 2 temperature regions downstream. Thus, from a smoke view-
$M0=70 vm point, a combustor may be considered to comprise two separate
z o n e s - - t h e primary zone, which governs the rate of soot forma-
UA=5 rn/s tion, and the intermediate zone (and, on modern high-tempera-
ture engines, the dilution zone also), which determines the rate
_PA' MPo (am) of soot consumption. The soot concentration actually observed
I0 0 0.76 (7.5) in the exhaust gas is an indication of the dominance of one
8 1.01 {10.0) zone over the other.
-- 0 1.27 (12.5)
Influence of Pressure. Problems of soot and smoke are
6 always most severe at high pressures. There are several reasons
0.6 0.8 1.0 1.2 1.4 for this; some derive from chemical effects, while others stem
from physical factors that affect spray characteristics and hence
the distribution of mixture strength in the soot-forming regions
Fig. 13 Influence of combustion pressure on CO emissions of the flame. For premixed kerosine/air flames it is found that
no soot is formed at pressures below 0.6 MPa and equivalence
ratios below 1.3.
of 287 K were around three times higher than the corresponding One adverse effect of an increase in pressure is to extend the
values at 298 K. A correlation of these data carried out by Hung limits of flammability, so that soot is produced in regions that,
(1993) yielded the following expression for calculating the ef- at lower pressures, would be too rich to burn (see Fig. 15).
fect of ambient air temperature on CO. It is considered valid Increased pressure also accelerates chemical reaction rates, so
for values of temperature up to 303 K. that combustion is initiated earlier and a larger proportion of
the fuel is burned in the fuel-rich regions adjacent to the spray.
COr/CO2ss = 1 - 0 . 0 6 3 4 ( T - 288) (5) With pressure atomizers, reduced spray penetration is one of
the main causes of smoke at high pressures. Thus whereas at
where COt = emissions of CO in ppm at 15 percent 02 at low pressures the fuel is distributed across the entire combustion
ambient temperature, T, and CO28s = emissions of CO in ppm zone, at high pressures it tends to concentrate in the soot-form-
at 15 percent 02 at 288 K. This equation should be used with ing region just downstream of the nozzle. Still another adverse
caution because it is likely to be very engine specific. Neverthe- effect of an increase in pressure is to reduce the cone angle of
less, it describes quantitatively what many other workers have the spray. This encourages soot formation, partly by increasing
observed, namely the high sensitivity of CO emissions to inlet the mean fuel drop size, but mainly by raising the mixture
air temperature. This small but important point needs to be
strength in the soot-forming zone.
borne in mind constantly when making comparative combustor The total effect of all these factors is that smoke emission
tests for CO emissions at different times. increases steeply with pressure, as shown in Fig. 16. Airblast
A further important factor influencing CO emissions is the atomizers are spared these problems because the fuel drops they
amount of liner wall-cooling air employed in the primary com- produce are always airborne. The distribution of fuel through
bustion zone. CO and UHC formed in primary combustion can the combustion zone is dictated solely by the liner airflow pat-
migrate toward the liner walls and become entrained in the tern, which is not susceptible to changes in pressure.
wall-cooling air. The temperature of this air is so low that all
chemical reactions are effectively "chilled." Thus the film-
cooling air emanating from the primary zone normally contains
high concentrations of CO and UHC. Unless these species are
subsequently entrained into the hot central core with sufficient
time to react to completion, they will appear in the exhaust gas.
Thus, reduction in film-cooling air by more efficient cooling


(or even ceramic liners) in the primary zone is effective in f COMBINES HIGH TEMPERATURE
Unburned Hydrocarbons. These include fuel that AND LOW OXYGEN CONTENT.
emerges from the combustor in the form of drops or vapor, as
well as the products of the thermal degradation of the parent BURNED
fuel into species of lower molecular weight. They are normally
associated with poor atomization, inadequate burning rates, the
chilling effects of film-cooling air, or any combination of these.
The reaction kinetics of UHC formation are more complex than

for CO formation, but generally it is found that those factors

that influence CO emissions also influence UHC emissions and Fig. 14 Soot-forming zone for conventional combustors

J o u r n a l of E n g i n e e r i n g f o r G a s T u r b i n e s a n d P o w e r OCTOBER 1995, Vol. 117 / 6 2 7

Downloaded From: on 01/27/2016 Terms of Use:

~ l , ~ , ~ 7 ~ LIMIT P3~ 0.54 MPo
~ o o
~ 4c

~( 0 0


10 ' I0 I 2~ 50
Fig. 17 Correlation of smoke number with smoke point for a TF 33
ally high quality of the data fit obtained with these and several
Fig. 15 Influence of pressure on flammability limits other aircraft combustors led Chin and Lefebvre to conclude
that smoke point is superior to hydrogen content as a correlating
parameter for soot-related combustion phenomena.
Influence of Fuel Type. Fuel properties can influence smoke
production in two ways: first by inducing the formation of local Oxides of Nitrogen. Some of the nitric oxide (NO) formed
fuel-rich regions, and second, by exerting variable resistance to in combustion can subsequently oxidize to NO2. For this reason
carbon formation. The former is controlled by physical proper- it is customary to lump NO and NO2 together and express results
ties, such as viscosity and volatility, which affect the mean drop in terms of oxides of nitrogen (NOx), rather than NO. At high
size, penetration, and rate of evaporation of the fuel spray, emission levels the predominant compound is nitric oxide,
whereas the latter relate to molecular structure. It is well estab- which is produced mainly by the oxidation of atmospheric nitro-
lished that smoking tendency increases with a reduction in hy- gen in high-temperature regions of the flame. The process is
drogen content and, in fact, hydrogen content is commonly used endothermic and it proceeds at a significant rate only at tempera-
in correlating rig and engine test data on various soot-related tures above around 1850 K. It can be produced by four different
parameters such as smoke emissions, flame radiation, and liner mechanisms, as discussed below.
wall temperature. However, Chin and Lefebvre (1993) have 1 Thermal NO. This is produced by oxidation of atmo-
shown that a better index of sooting tendency is the ASTM spheric nitrogen in the postflame gases. Most of the proposed
smoke point, which is obtained experimentally by burning the reaction schemes for thermal NO utilize the extended Zeldovich
test fuel in a wick lamp and slowly increasing the height of mechanism:
the flame until it begins to smoke. The height of the flame in
millimeters is the smoke point; the higher this is, the lower is N2 + O = NO + N
the tendency of the fuel to soot formation.
N + 02 = NO + O
The correlations shown in Figs. 17 and 18 were obtained
from measurements of smoke number carried out on Pratt and N + OH = NO + H
Whitney TF33 and F100 combustors, respectively. The gener-
NO formation is found to peak on the fuel-lean side of stoichio-
metric. This is a consequence of the competition between fuel
200 and nitrogen for the available oxygen. Although the combustion
temperature is higher slightly on the rich side of stoichiometric,
the available oxygen is then consumed preferentially by the
I00 fuel. For even richer mixtures, a further reduction of NO may

0d 30 P:5"H2-LI8 MPo



IC ! I I I I I
I0 20 50
Fig. 18 Correlation of smoke number with smoke point for an F 100
Fig. 16 Data illustrating the effect of pressure on exhaust smoke combustor

628 / Vol. 117, OCTOBER 1995 Transactions of the ASME

Downloaded From: on 01/27/2016 Terms of Use:

P - 1.44 MPa (14.2 aim) FUEL=DF 2 _TA, K
T - 6 5 0 - 730 K
PA= 1.01 MPo 0 57'3
50 SMD =110 p.rn 0 473


20 45C

~. lO


06 0.8 ID 1.2
11/050 I
1600 I
1700 I
1800 I
1900 I
Fig. 20 Influence of inlet air temperature on nitric oxide emissions

Fig. 19 Dependence of NOx on flame temperature for liquid and gaseous

1 Thermal NO formation is controlled largely by flame tem-
2 Little NO is formed at temperatures below around 1850 K.
3 For conditions typical of those encountered in conventional
occur if sufficient hydrocarbon radicals and adequate residence gas turbine combustors (high temperatures for less than a
time are available. few milliseconds), NO increases linearly with time but does
Figure 19 illustrates the exponential dependence of NOx on not attain its equilibrium value.
flame temperature for both gaseous and liquid fuels. It is based 4 For very lean premixed combustors (~b < 0.5), NO forma-
on experimental data (not shown in the figure) obtained by tion is largely independent of residence time.
Snyder et al. (1994) in their studies on the combustion perfor-
mance achieved when using a tangential entry lean-premixed
fuel nozzle. Data similar to those shown in Fig. 19 have been IO.O
obtained by many different workers using many different types
of combustion systems. Of special interest in this figure is that
the well-known difference in NOx emissions between liquid and
gaseous fuels diminishes with increase in flame temperature,
becoming negligibly small at the highest levels of temperature. m
With liquid fuels there is always the potential for near-stoichio- 59
metric combustion temperatures, and consequently high NOx
formation, in the region surrounding the fuel drops, even though
the average equivalence ratio throughout the combustion zone
is appreciably less than stoichiometric. With increase in equiva-
lence ratio, the ambient flame temperature becomes closer to t.tJ
the stoichiometric value, so that local conditions around the fuel Z
drops have less influence on the overall combustion process ~ 2.0
and the NOv emissions begin to approximate those produced by O
gaseous fuels.
As NO emissions are very dependent on flame temperature, a.
an increase in inlet air temperature would be expected to pro- ..~
duce a significant increase in NO, and this is confirmed by the "~
results shown in Fig. 20 from Rink and Lefebvre (1989b). This l.O
figure contains data for a mean fuel drop size (SMD) of 110/zm,
but similar results were obtained when the SMD was reduced to O
30/.zm. Z
Combustor residence time can also influence NOx emissions,
as shown in Fig. 21, which contains results obtained by Ander-
son (1975) when using a premix-prevaporize combustor burn- 0.5
ing gaseous propane fuel. It shows that NOx emissions increase 0.4
with increase in residence time, except for very lean mixtures
(th = 0.4), for which the rate of formation is so low that it
becomes fairly insensitive to time. Similar results showing the
insensitivity of NOx formation to residence time in lean pre-
mixed combustion have been obtained by Leonard and Steg-
maier (1994) and Rizk and Mongia (1993a). These findings
have important practical implications to the design of lean pre- ,.b ,£ 2!o
mixed combustors. RESIDENCE TIME,ms
The key points regarding thermal NO may be summarized
as follows: Fig. 21 Effect of residence time on NO. in a premixed fuel-air system

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 629

Downloaded From: on 01/27/2016 Terms of Use:

2 Nitrous Oxide Mechanism. According to Nicol et al.
(1992) this mechanism is initiated by the reaction J /
30- P.ESsuAE y /
N2 + O = N20 ~ . " a.1 J / c /~ ¢ p
and the N20 (nitrous oxide) formed is then oxidized to NO a o181 821 j ~/ /
mainly by the reaction
N20 + O = N O + NO
but also by the reactions
N20 + H = N O + N H
NzO + CO = N O + NCO
3 Prompt NO. Under certain conditions, NO is found very I I I I I
early in the flame r e g i o n - - a fact that is in conflict with the °o.~o 0.65 0.70 0.75 0.80 0.8S 0.90
idea of a kinetically controlled process. According to Nicol et EQUIVALENCE RATIO
al. the initiating reaction is Fig, 22 Data illustrating the effect of pressure on NOx formation

N2 + C H = H C N + N
The balance of the prompt NOx mechanism involves the oxi- Such extrapolation could be carried out with confidence if the
dation of the HCN molecules and N atoms. Under lean-pre- relationship between NOx and pressure were accurately known.
mixed conditions, the HCN oxidizes to NO mainly by a se- Unfortunately, the experimental data obtained on different com-
quence of reactions involving HCN - , CN --* NCO ~ NO. The bustor types are conflicting in this regard. They vary from no
N atom reacts mainly by the second Zeldovich reaction. effect of pressure on NOx to quite significant increases in NOx
Of special interest and importance is the influence of pressure with increase in pressure.
on prompt NO formation because this could be the main contrib- For conventional combustors it is generally found that NOx
utor to the NO emissions produced in lean premixed combustion c~ pn, where n has values ranging from around 0.5 to around
(Correa, 1991 ). There are, however, few data available on this 0.8. Laboratory burners can also exhibit a strong effect of pres-
effect. Fennimore's (1971) pioneering study of prompt NO in sure on NOx. For example, the results of Maughan et al.. (1994)
ethylene-air flames over a range of pressures from 1 to 3 atm from a well-mixed combustor supplied with natural gas fuel
concluded that prompt NO ~ p0.5. Later work by Heberling showed that raising the combustor outlet temperature from 1227
(1976) over a much wider range of pressures from 1 to 18 atm to 1310 K caused n to increase from 0.38 to 0.51. Maughan et
showed that prompt NO was independent of pressure. Altemark al. regard this result as evidence that the lowest NOx levels
and Knauber (1987) also concluded that NOx is independent of result from the nitrous oxide and prompt mechanisms which
pressure for equivalence ratios below 0.6. The practical implica- dominate at low temperatures and which are independent of
tions of these findings are discussed below. pressure, whereas the higher NOx levels associated with higher
combustion temperatures are due primarily to thermal NOx
4 Fuel NO. If fuels contain organically bonded nitrogen,
which exhibits a square root dependence on pressure.
then some of this nitrogen will eventually form the so-called
These results and conclusions are fully consistent with those
"fuel NO." The percentage of nitrogen undergoing this change
obtained by Correa and co-workers (Leonard and Corea, 1990;
increases only slowly with increasing flame temperature. Light
Correa, 1991 ). These workers studied turbulent premixed meth-
distillate fuels contain small amounts of organic nitrogen, less
ane-air flames using an uncooled perforated plate burner op-
than 0.06 percent, but the heavy distillates may contain as much
erating at pressures from 1 to 10 atm, inlet air temperatures
as 1.8 percent. Thus, depending on the degree of nitrogen con-
from 300 to 615 K, and equivalence ratios from 0.5 to 0.9.
version, the fuel NO can represent a considerable proportion of
Their modeling featured a stirred reactor for flame stabilization
the total NO (Merryman and Levy, 1975).
followed by a plug flow reactor and a kinetic scheme which
Nicol et al. (1992) have examined analytically the relative
included thermal and prompt NO. The results confirmed that
contributions of the various mechanisms discussed above to the
the low temperatures of lean flames preclude significant forma-
total NOn emissions produced by a lean-premixed combustor
tion of NO by the thermal mechanism. At temperatures below
burning methane fuel, for which, of course, the fuel NO is
1800 K the prompt mechanism appears to be dominant. The
zero. The results of their study showed that at relatively high
implication of these results to the effect of pressure on NO2
temperatures of around 1900 K, and equivalence ratios of
formation is well illustrated in Fig. 22, which contains some of
around 0.8, the contributions are about 60 percent thermal, 10
the experimental data of Correa et al. and which highlights their
percent nitrous oxide, and 30 percent prompt. With reductions
conclusions in regard to the influence of flame temperature on
in temperature and equivalence ratio, the contributions made
the pressure dependence of NO~ formation. In this and in other
by nitrous oxide and prompt NO increase significantly until, at
figures contained in this paper, the values of NO and NOx quoted
a temperature of 1500 K and an equivalence ratio of around
have been corrected to the values that would have been obtained
•0.6, the relative contributions to the total NOx emissions become
if the combustion air had contained 15 percent oxygen. The
5 percent thermal, 30 percent nitrous oxide, and 65 percent
following expression is used to make this correction:
prompt. These results clearly have great importance to the de-
sign of ultralow NOx lean-premixed combustors. (NOx)15%O2 = (NOx) × 5.9/(20.9 - 02 .......) (6)
Influence of Pressure. The influence of pressure on NOx where NOx concentrations are expressed in ppm by volume on
formation is of special importance due to the continual trend a dry basis and 02 content is expressed as a percentage by
toward engines of higher pressure ratio to meet the need for volume on a dry basis.
lower fuel consumption. Combustor testing at high pressures is The purpose of this parameter is partly to remove ambiguity
extremely expensive and it would therefore be highly conve- when comparing different sets of experimental data, but mainly
nient to carry out combustor development at lower levels of to indicate that combustors burning less fuel are expected to
pressure and then extrapolate the results obtained to high levels produce less NOx. Figure 22 shows that NOx is independent
of pressure where NOx emissions attain their highest values. of pressure in the leanest premixed flames. Increase in flame

630 / Vol. 117, OCTOBER 1995 Transactions of the ASME

Downloaded From: on 01/27/2016 Terms of Use:

temperature, corresponding to an increase in equivalence ratio, 2500
causes the pressure exponent to increase until, in the near-stoi-
chiometric region, it attains the value of 0.5, corresponding to LEAN O NON FILM COOLED
NO formation by the thermal mechanism. BLOWOUT E] FILM COOLED
2000 -
Additional evidence to support the argument that NOr forma-
tion in well-mixed, low-temperature flames is largely indepen-
dent of pressure has been provided by Leonard and Stegmaier ui
(1994) and Steele et al. (1995). Their experiments covered ~ 1500-
ranges of pressure from 1 to 30 atm and 1 to 7.1 atm, respec- zo
tively. Leonard and Stegmaier found little or no effect of pres-
sure, while the results obtained by Steele et al. using a lean, z 1000-
premixed, high-intensity combustor showed a neutral or even o
slightly negative effect of pressure on NOn. Comparatively little ¢
is known about the pressure dependence of NOx formation in
fuel-rich flames. Rizk and Mongia (1991b) performed a three- P - 110 kPa ~ I
dimensional analysis to examine the influences of pressure and
residence time on NOx formation in the rich zone of a rich/ T - 750 K_i I I
quench/lean (RQL) combustor. Their predictions indicate that 0
the value of the pressure exponent n varies with rich-zone equiv- 1600 1700 1800 1900 2000
alence ratio according to the relationship: PRIMARY COMBUSTION TEMPERATURE, K
n = 116.5 exp - (q5/0.222) (7) Fig. 23 Effect of eliminating hot-side film cooling on CO
For a typical rich zone equivalence ratio of 1.4 this equation
gives a value for n of 0.68.
adequate burning rates or so fuel-rich that there is insufficient
02 to convert all the CO produced into CO2.
Pollutant Reduction in Conventional C o m b u s t o r s
As mentioned earlier, an effective means of reducing CO and
Although it might reasonably be argued that conventional UHC is by using less liner wall-cooling air, especially in the
combustors are no longer of any real technical interest, they do primary zone. Figure 23 shows the effect of replacing a conven-
nevertheless constitute the big majority of combustors now in tional film-cooled wall with a non-film-cooled wall in the pri-
service. Furthermore, most of our knowledge of the key factors mary zone of a Rolls-Royce Industrial RB211 low-emissions
governing pollutant formation in continuous flow combustion combustor when operating at atmospheric pressure. CO is sig-
systems, which is now being applied to the design and develop- nificantly reduced (at 1850 K from 1500 ppm to 700 ppm)
ment of ultralow-NOx combustors, was acquired from experi- while the lean blowout temperature is lowered by 110 K (Willis
ence gained on what are now called "conventional" combus- et al., 1993).
tors. The development of new materials and methods of liner wall
The main factors controlling emissions from conventional construction, which allow the liner to operate at higher metal
combustors may be considered in terms of: temperatures, along with the development of new methods of
1 The primary-zone temperature and equivalence ratio. wall cooling, which require much less cooling air, such as effu-
sion and transpiration cooling, make a very direct and significant
2 The degree of homogeneity of the primary-zone combustion
process. contribution to the reduction of CO and UHC emissions.
3 Residence time in the primary zone. Another method of reducing these emissions is by improved
4 Liner-wall quenching characteristics. fuel atomization. Figure 24 shows that improvements in atom-
ization quality, which enhance evaporation rates, yield apprecia-
With liquid fuels, spray properties are also of prime importance. ble reductions in CO, except at low equivalence ratios where
In considering practical design methods for pollutant reduc- the influence of fuel drop size is less pronounced. This is be-
tion, attention is focused first on individual pollutant species. cause the increase in evaporation rates arising from reductions
It will become clear, however, that with conventional combus- in mean drop size offer little advantage at low equivalence
tors a great deal of compromise is involved in design, not only ratios where burning rates tend to be limited more by chemical
between one species and another, but also among the many reaction rates than by evaporation rates. Figure 25 shows that,
other performance requirements, such as lean blowout limits, in common with CO, UHC emissions are greatly reduced by
pattern factor, and size. reductions in Sauter mean diameter (SMD). (Note the log scale
in this figure.)
Carbon Monoxide and Unburned Hydrocarbons. The The main effect of mean drop size on emissions stems from
presence of these species in the exhaust gases is a manifestation its strong influence on the volume required for fuel evaporation.
of incomplete combustion. Thus all approaches to CO and UHC At low power operation, where CO and UHC emissions attain
reduction are based on a common philosophy, which is to raise their highest concentrations, a significant proportion of the total
the level of combustion efficiency. An effective method of combustion volume Vc is occupied in fuel evaporation. Conse-
achieving this is by redistributing the airflow to bring the pri- quently, less volume is available for chemical reaction. Under
mary-zone equivalence ratio closer to the optimum value of these conditions, any factor that influences evaporation rates,
around 0.8. A higher equivalence ratio (up to around 1.0) would such as fuel volatility and mean drop size, will have a direct
increase burning rates even further, but it would not necessarily effect on the volume available for chemical reaction (V,. - Ve)
yield lower emissions of CO and UHC due to the dearth of and, therefore, on the emissions of CO and UHC.
oxygen, which these species need in order to convert to CO2
and H20. An equivalence ratio slightly less than stoichiometric Smoke. The main factors governing smoke emissions are
is usually regarded as optimal for minimum CO and UHC. combustor inlet air temperature, pressure, and fuel spray charac-
Good fuel-air mixing in the primary zone is essential for teristics. The influence of inlet air temperature is complex be-
low CO and UHC. When operating at the optimal equivalence cause an increase in this parameter serves to accelerate both the
ratio of around 0.8, poor mixing could produce localized regions soot-forming and the soot-burnout processes; the net result is
in which the mixture strength is either too fuel-lean to provide usually a reduction in smoke. Smoke problems are most severe

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 631

Downloaded From: on 01/27/2016 Terms of Use:



PA" 1.01 MPa 0 30 TA=ST3K / / /
& 70 UA=5 m/s / L/
3.0 PA=1'52
200 E] 110

E 2.C
80 f o_
§ ,o
40 / / / ,SMD, ~ m
u) 0.8
I ip
I dl
/ // IZ1II0
20 0.6
Q5 >.,

I I | I I I I I I
O8 Q9 I0 I.I 1.2 13
10 0.6 0.8 1.0 1.2 1.4 EQUIVALENCE RATIO
EQUIVALENCE RATIO, Fig. 26 Influence of mean drop size on smoke emissions

Fig. 24 Influence of mean drop size on CO emissions

in pressure also accelerates chemical reaction rates, so that com-

at high pressures. There are several reasons for this, most of bustion is initiated earlier and a larger proportion of the fuel is
which derive from chemical effects. With liquid fuels there are burned in the fuel-rich regions adjacent to the spray.
additional physical factors, which affect spray characteristics In practice the elimination of exhaust smoke is basically a
and hence also the distribution of mixture strength in the soot- matter of preventing the occurrence of fuel-rich pockets in the
forming regions of the flame. flame. Injecting more air into the primary zone is always bene-
One important consideration is that burning limits widen with ficial, especially if accompanied by more thorough mixing. Un-
pressure. Thus, at high pressures soot is produced in regions fortunately, this approach is somewhat limited in scope, owing
which, at lower pressures, would be too rich to burn. Increase to the adverse effect of an increase in primary-zone air on
ignition and stability limits and on CO and UHC emissions at
4.0, The influence of mean drop size on smoke emissions derives
FUEL= FL 0653 SMD~ ,u.._m. from the manner in which each individual droplet evaporates.
Conventional gas turbine fuels are generally multicomponent
TA=573 K 0 30 mixtures of various petroleum compounds, each of which has
UA =5 m/s ~ "70 different physical and chemical properties. As evaporation pro-
~ 3 . 5 - PA =1'01 MPo 0 I10 ceeds, the composition of a multicomponent fuel drop changes
by a process of simple batch distillation. The more volatile
constituents of the fuel drop vaporize first and the concentration
of the higher-boiling fractions in the remaining liquid phase
=o 30 During the droplet evaporation process the droplet receives
heat from the surrounding combustion gases. If the droplet is
small in diameter it is usually fully evaporated before any sig-
D nificant temperature rise can occur. However, with droplets of
large diameter, the heat flowing radially inward during droplet
8 2.5 evaporation has sufficient time to raise the droplet temperature,
$ thereby promoting the formation of carbon and soot in the
"heavy ends" located at the center of the droplet.
The influence of fuel drop size on exhaust smoke has been
investigated by Rink and Lefebvre (1989a, b) using the tubular
z.o combustor described above, supplied with a kerosine fuel. Their
results for a combustion pressure of 1.27 MPa (12.5 atm) are
shown in Fig. 26. This figure shows that particulates are dimin-
ished by reductions in mean drop size. For example, at the
highest equivalence ratios, reducing the mean drop size from
0.4 0.6 0.8 IO 1.2 110 to 30/zm effectively halves the particulate concentration.
The importance of atomization quality to soot formation and
EOUIVALENCE RATIO smoke stems from the fact that, as the fuel spray approaches
Fig. 25 Influence of mean drop size on UHC emissions the flame front, heat transmitted from the flame starts to evapo-

632 / Vol. 117, OCTOBER 1995 Transactions of the ASME

Downloaded From: on 01/27/2016 Terms of Use:

70 reason for this is not just better atomization, although this is
very significant at high combustion pressures where smoke lev-
P&W JT9D-70A els attain their highest values, but because the airblast atomiza-
60 TAKEOFF tion process virtually guarantees good mixing of air and fuel
drops prior to combustion. The marked reduction in exhaust
smoke that can be achieved with airblast atomizers is illustrated
50 in Fig. 28, in which smoke levels are plotted against overall
combustor air/fuel ratio for a dual-orifice nozzle and a prefilm-
UJ ing airblast atomizer at the same operating conditions. The air-
40- blast and dual-orifice atomizers exhibit opposite trends in regard
Z to the variation in smoke output with air/fuel ratio. This is
I..LI because the superior atomization and mixing performance of
0 30- the airblast atomizer ensures that burning is virtually complete
within the primary zone. Thus, any increase in overall air/fuel
ratio leads automatically to improved aeration of the combustion
20 7. process and hence to a decrease in smoke. In contrast, a decrease
in fuel flow through the pressure atomizer reduces the atomiza-
tion quality and raises the fuel concentration in the soot-forming
region just downstream of the nozzle.
From an emissions viewpoint, another important advantage
of the airblast atomizer is that atomization quality is high over
I I I I I I ~ 'I" Iv
2 4 6 8 10 12 14 16 18 the entire operating range from idle to full power. With a duai-
orifice injector, owing to the interaction of the pilot and main
sprays, there is always a range of fuel flows, starting at the
Fig. 27 Control of exhaust smoke through fuel injector and swirler air- point where the main fuel is first admitted, over which atomiza-
flows tion quality is poor. Since atomization quality has a direct bear-
ing on CO and HC emissions, it follows that the use of dual-
orifice atomizers must inevitably aggravate the emissions prob-
rate the drops. The smallest droplets in the spray have time to lem over a significant range of engine operation. With the
evaporate completely ahead of the flame front, and the resulting piloted-airblast injector, there is no physical interference be-
fuel vapors then mix with the combustion air and burn in the tween the pilot and main sprays and, because no fuel is intro-
manner of a premixed flame. The largest drops in the spray do duced into the airblast section until the air velocity through it
not have time to evaporate fully and mix completely with air. has attained its normal operating level, atomization quality is
Instead, they burn in the mode of fuel-rich diffusion flames. high throughout the entire range of engine operation.
Clearly, any increase in mean drop size will increase the propor-
tion of large drops in the spray. This, in turn, will raise the Oxides of Nitrogen. In any attempt to reduce NOx, the
proportion of fuel burned in diffusion-type combustion, as op- prime goal must be to lower the reaction temperature. The sec-
posed to premixed combustion. ond objective should be to eliminate hot spots from the reaction
It is important to bear in mind that soot formation is nearly zone, as there is little point in achieving a satisfactorily low
always the result of local fuel-rich combustion. In general, ex- average temperature if the reaction zone contains local regions
haust smoke decreases with mean drop size, but if improved of high temperature in which the rate of NOx formation remains
atomization leads to a reduction in spray penetration, as occurs high. Finally, the time available for the formation of NOx should
with all types of pressure atomizers, the smoke output may be kept to a minimum.
actually increase. In fact, reduced spray penetration is one of Practical approaches to low NOx in conventional combustors
the main causes of smoke on high-pressure-ratio engines fitted include the addition of more air into the primary combustion
with dual-orifice atomizers. zone to lower the flame temperature, increase in liner pressure
Another adverse effect of an increase in pressure is to reduce drop to promote better mixing and thereby eliminate hot spots
the cone angle of the fuel spray. This encourages soot formation, from the combustion zone, and reduction in combustor resi-
partly by increasing the mean drop size, but mainly by raising dence time. Unfortunately, reductions in flame temperature and
the mixture strength in the soot-forming zone. residence time lead to increased output of both CO and UHC.
The design of the fuel injector and, in particular, the degree In fact, as a generalization, it can be stated that any change in
of premixing of fuel and air prior to combustion, have a very operating conditions or combustor configuration that reduces
large influence on whether or not a given combustor will pro-
duce significant amounts of smoke. The relatively low smoke
emissions from the vaporizer systems employed on some Rolls- 25

Royce engines is not due to prevaporization of the fuel but DUAL
rather to the premixing of fuel and air which occurs within the ¢n20 ORIFICE
vaporizer tubes. Z
Alleviating soot formation and smoke by fuel-air mixing is
only fully effective if sufficient air is used. This is well illus- w 15
trated in Fig. 27 from Sturgess et al. (1993), which shows how
smoke was drastically reduced in a P&W JT9D-70 combustor ¢n I0
when operating at takeoff conditions by the addition of more
air through the fuel injector and air swirler. The injection of air
through these components is particularly effective in reducing
smoke because it all flows into the primary combustion zone, ' 9--A'RBLAST,
which is usually deficient in oxygen and therefore has a high 0
140 160 180
tendency toward soot formation.
The advantage of airblast atomizers over dual-orifice pressure
atomizers in regard to smoke emissions is well established. The Fig. 28 Influence of fuel injector type on exhaust smoke

J o u r n a l of E n g i n e e r i n g for G a s T u r b i n e s a n d P o w e r OCTOBER 1995, Vol. 117 / 633

Downloaded From: on 01/27/2016 Terms of Use:

CO/HC NOx prevaporize combustors whose success relies largely on the
COMBUSTION EFFICIENCY SMOKE elimination of all fuel drops from the combustion zone.
W a t e r Injection. As NOx formation is exponentially de-
pendent on temperature, an obvious way of reducing NOr emis-
POWER POWER sions is by lowering the temperature of the combustion zone.
Additional air is effective, but can only be used sparingly be-
t , t cause it raises the primary zone velocity, which has an adverse
effect on both ignition and stability performance. An alternative
INCREASE COMBUSTIONVOLUME DECREASE COMBUSTIONVOLUME approach is to introduce a heat sink, such as water or steam,
INCREASE RESIDENCE TIME DECREASE RESIDENCE TIME into the combustion zone. This is clearly unsuitable for aero-
engines but is a practical proposition for large stationary en-
Fig. 29 Illustration of tradeoffs between various aspects of combustion gines, especially if large amounts of water or steam are avail-
performance able. In some cases the water or steam is injected directly into
the flame, either through a number of separate nozzles located
at the head end of the combustor or through holes that are
integrated into the fuel nozzle (Hilt and Waslo, 1984). Alterna-
NOx tends also to exacerbate the problems of CO and UHC,
tively, the water injection may take place upstream of the com-
and vice versa. bustor liner, usually into the air stream, which subsequently
Design modifications for reducing pollutant emissions gener-
flows into the combustion zone through the main air swirler.
ally have an adverse effect on other performance parameters
This method ensures good atomization, because the smaller
such as ignition and altitude relight, lean blowout, and outlet
droplets are carried by the air flow through the swirler into the
temperature distribution (pattern factor), as illustrated in Fig.
combustion zone, while the larger drops impinge on the swirler
29 from Bayle-Labour6 ( 1991 ). Thus, in the design of conven-
vanes where they form a thin liquid film, which is airblast
tional gas turbine combustors, the end result must inevitably be
atomized as it flows over the downstream edge of the vane
a compromise of some kind in regard to both emissions and
(McKnight, 1978).
other aspects of combustion performance.
When steam is used to reduce NOx emissions, it may also be
The manner and extent to which oxides of nitrogen are influ-
injected directly into the combustion zone or into air, which
enced by the sizes of the fuel droplets in the spray is very
subsequently flows into the combustion zone. In some installa-
dependent on equivalence ratio. This aspect was addressed in
tions the steam is injected into the compressor discharge air.
the experimental studies carried out by Rink and Lefebvre
This method is simple but inherently wasteful, because only
(1989a, b) using the continuous flow combustor referred to
about 40 percent of the steam actually flows into the combustion
above in which the mean drop size could be varied and con-
zone. This may be only a minor consideration if excess steam
trolled independently of other operating variables. The data pre-
is available (Hilt and Waslo, 1984).
sented in Fig. 30 show that NO emissions increase with increase
The effectiveness of water and steam for reducing NOn has
in mean drop size, especially at low equivalence ratios. At
been demonstrated by many workers. According to Hung
first sight this may seem surprising, because at high pressures
(1974) the relationship between NOx reduction and water/fuel
evaporation rates are so fast that even for the larger drops the
mass ratio, X, can be expressed as
time required for their evaporation is small in comparison with
the total residence time of the combustion zone. However, an wet N O J d r y NOx = exp - (0.2X 2 + 1.41X) (8)
increase in SMD means that a larger proportion of the total
number of fuel drops in the spray are capable of supporting This relationship was found to apply to both liquid and gaseous
"envelope" flames. These envelope flames, which surround the fuels. It shows, for example, that equal mass flow rates of water
larger drops, bum in a diffusion mode at near-stoichiometric and fuel (for which X = 1) yield an 80 percent reduction in
fuel/air ratios, giving rise to many local regions of high temper- NOx. Very similar results were obtained for both gaseous and
ature in which NOr is formed in appreciable quantities. Reduc- liquid fuels by Claeys et al. (1993) on the General Electric
tion in mean drop size impedes the formation of envelope MS7001F gas turbine.
flames, so that a larger proportion of the total combustion pro- Equation (8) should not be regarded as having universal
cess occurs in what is essentially a premixed mode, thereby application. For example, Wilkes (1980) has shown that water
generating less NOx. Even if no envelope flames are present, injection is much less effective with fuels containing fuel-bound
with increasing drop size a larger proportion of the fuel bums nitrogen. The main effect of water addition is to reduce thermal
in the fuel-rich regions created in the wakes of the moving
drops. Although in theory combustion can take place at any
equivalence ratio within the flammability limits, it tends to occur
preferentially at the stoichiometric value, i.e., at the maximum
~ ~ _ T A - 573 K
temperature, thereby producing high levels of NOx. This hy- 16
PA" 1.27 MPa
pothesis serves to explain why NOx emissions increase with
SMD for lean mixtures. However, as the overall equivalence 14 SMD~
ratio increases toward unity the local fuel/air ratio adjacent to o 30--
the fuel drops approaches the premixed value. According to this "~" 1 2 A 70
hypothesis, mean drop size should have no influence on NOx
emissions for stoichiometric mixtures, and this is generally con-
firmed by the results shown in Fig. 30. This figure is important ZOO"lif
because it demonstrates that even at low equivalence ratios
where the average combustion temperature is so low that only
negligible amounts of NO should, in theory, be formed, the
presence of fuel drops in the combustion zone must necessarily I ~" i l I I I I
give rise to conditions in which combustion can and does pro- 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3
ceed at near-stoichiometric equivalence ratios regardless of the EQUIVALENCE RATIO,
average equivalence ratio in the combustion zone. This, of
course, is the rationale for the various types of lean, premix, Fig, 30 Influence of mean drop size on NO emissions

634 / Vol. 117, OCTOBER 1995 Transactions of the ASME

Downloaded From: on 01/27/2016 Terms of Use:

NOx although it does also slightly reduce prompt NO. This 120 30
implies that water injection is most effective when combustion
takes place at high pressures and temperatures where thermal TEMPERATURE i
NOx production is high, and is less effective at low pressures _ _.p RANGE FOR L.._ .
and temperatures where a larger proportion of the total NOx is LOW F I
formed via the prompt mechanism. The key point is that water ~> EMISSIONS I /
and/or steam injection always reduces NOx, but the extent of
-20 i
the reduction depends on combustor operating conditions and uJ
fuel type.
As steam is a less effective diluent than water, the reductions 50 I -15L-
in NOx achieved with steam injection tends to be less dramatic Z
than when water is used. For example, Hilt and Waslo (1984)

i =
have reported NOx reductions of around 60 percent for a steam/
fuel mass ratio of unity on two GE industrial engines burning -10~
natural gas.
Although both water and steam injection are very effective
in reducing NOx emissions, and have been used on stationary - 5
engines that operate at near-constant load conditions since the
early 1970s, they do have a number of drawbacks. White et al.
(1982) have reported an increase in capital cost of 10 to 15 21000
U.S. dollars per kW and an increase in fuel consumption of 2
to 3 percent. This additional fuel is needed to heat the water to PRIMARY TEMPERATURE, K
combustion temperature, although power output is enhanced
due to the additional mass flow through the turbine. The water 1000 K ~ TEMPERATURE RANGE OF : 2500 K
must be of high purity to prevent deposits and corrosion in the
hot sections downstream of the combustor. The treatment of Fig. 31 Influence of primary-zone temperature on CO and NOx emis-
this water is expensive and requires a separate plant based on sions
reverse osmosis and de-ionization. Users' experience with water
injection has shown a significant increase in inspection and
hardware maintenance. There are, therefore, practical limits to weighs 111 tons. Despite these drawbacks, the method is quite
the amount of water or steam that can be injected into the widely used.
combustor. The deterioration in combustion performance aris-
ing from water-steam injection is manifested as increases in the Exhaust Gas Recirculation. It has been known for many
levels of CO and UHC emissions and by increases in combustor years that recirculation of cooled exhaust products into the inlet
pressure oscillations. These oscillations can become amplified of a gas turbine combustor would reduce the emissions of NOx.
by coupling with the combustion process, and cause deteriora- The practical feasibility of this method of NOx reduction has
tion of combustor hardware. been investigated by Wilkes and Gerhold (1980) who found
The drawbacks associated with water and steam injection that significant reductions (50 percent) could be achieved with
have encouraged the development of the so-called "dry low recirculation rates of 20 percent or less at baseload conditions.
NOx" combustors, i.e., combustors that can meet the emission The major thermal effect stems from the reduced concentration
goals without having to resort to diluent injection. of oxygen in the inlet air, but there is also a secondary effect
Selective Catalytic Reduction ( S C R ) . This is a method due to the higher heat capacity of this air with increased H20
for converting NOx in a gas turbine exhaust stream into molecu- and CO2 content.
lar nitrogen and water vapor by injecting ammonia into the The main advantage of the method is that little or no combus-
stream in the presence of a catalyst. Exhaust gases first pass tor development is required and standard production combustors
through an oxidation catalyst and are then mixed with ammonia can be used. Its main drawback lies in the need for an intercooler
before entering the SCR catalyst. The oxidation catalyst re- between the exhaust and inlet. This virtually rules it out for
moves the CO and UHC emissions by oxidizing them to CO2 simple gas turbines, but application to combined cycle plants
and H20. To reduce NOr emissions, ammonia is injected in a offers more promise due to the substantially lower exhaust gas
manner designed to achieve intimate mixing with the exhaust temperatures. Another drawback is that only very clean fuels
stream. After mixing, the exhaust gases pass over a base metal can be used to avoid problems of fouling and contamination.
catalyst which results in the selective reduction of NOx to form
N2 and H20. The principal reactions are
Pollutant Reduction by Control of Flame Tempera-
6NO + 4NH3 ~ 5N2 + 6H20 ture

6NO2 + 8NH3 ~ 7N2 + 12H20 From the foregoing discussion it is abundantly clear that the
main parameter governing pollutant emissions is the tempera-
SCR requires that the temperature of the exhaust stream be ture of the combustion zone. With conventional combustors this
within a fairly narrow range from 560 to 670 K, and so is can range from 1000 K at low power operation to 2500 K at
restricted to systems in which the exhaust gas flows into a heat high power operation, as indicated in Fig. 31. This figure also
recovery device, usually a steam generator (Davis and Washam, shows that too much CO is formed at temperatures below
1989). A major problem with this method is the requirement around 1670 K while excessive amounts of NOx are produced
for a control system that feeds the requisite amount of ammonia, at temperatures higher than around 1900 K. Only in the fairly
and the need for a continuous monitoring system that can give narrow band of temperatures between 1670 and 1900 K are the
the feedback to the ammonia supply mechanism under differing levels of CO and NOx below 25 and 15 ppm, respectively.
load conditions. Another problem is the size and weight of the The underlying principle of the various approaches toward low-
equipment. According to Davis and Washam, for an 83 MW emission combustors described below is that of maintaining the
MS7000 gas turbine an SCR designed to remove 90 percent of combustion zone (or zones) within a fairly narrow band of
the NOx from the exhaust stream has a volume of 175 m 3 and temperatures over the entire power range of the engine.

Journal of Engineering for Gas Turbines and P o w e r OCTOBER 1995, Vol. 117 / 635

Downloaded From: on 01/27/2016 Terms of Use:

Variable Geometry. The notion of variable-geometry
combustors is by no means new. Long before the problems of
pollutant emissions were recognized, many different schemes
for incorporating variable geometry into gas turbine combustors
were considered, usually as a means for improving the altitude
relight performance of aircraft engines. These schemes seldom
came to fruition because designers, while appreciating the ad-
vantages to be gained, were reluctant to accept the mechanical
complexities involved. However, the opportunities it offers for
emissions reduction has resulted in a revival of interest in the
use of variable geometry, especially for nonaeronautical appli-
Many practical forms of variable-geometry combustors have
been designed. A fully variable system is one in which large
quantities of air are admitted at the upstream end of the combus- ~ - ~ ~- UNFUELED
tion liner at max power conditions to lower the primary-zone FUELED
temperature and provide adequate film-cooling air. As engine
power is reduced, an increasing proportion of this air is diverted
to the dilution zone to maintain the primary-zone temperature
within the low-emissions "window" shown in Fig. 31. Practical
ways of achieving this variation in airflow distribution include
the use of variable-area swMers to control the amount of air
flowing into the combustion zone; see, for example, Bayle-
Labour6 (1991) and Micklow et al. (1993), variable air open-
ings into the dilution zone (Roberts et al., 1982; Sasaki et al.,
1991), or a combination of these.
The drawbacks to all forms of variable geometry systems
include complex control and feedback mechanisms, which tend
to increase cost and weight and reduce reliability. Problems of
achieving the desired temperature pattern in the combustor ef-
flux gases could also be encountered, especially if the liner
pressure drop is allowed to vary too much. The incentive for
surmounting these practical problems is that variable geometry
has the potential for reducing simultaneously all the main pollut-
ant species without sacrificing other aspects of combustion per-
formance. It also has several other advantages, for example, as
the combustion temperature is always maintained above a cer-
tain minimum value of around 1670 K, chemical reaction rates
are always relatively high. This enables the combustion zone
to be made smaller, with consequent advantages in terms of
reductions in combustor size and weight. For aircraft applica-
tions, variable geometry also has the potential for wide stability Fig. 32 Illustration of the use of selective fuel injection
limits and improved altitude relight performance.
To be fully effective, variable geometry combustors should
ideally be used in conjunction with premix-prevaporize fuel
injection systems. Only in this way is it possible to avoid the mon use, not only reduces CO and UHC emissions but also has
local high-temperature, high-NOx-forming regions, created by the added advantage of extending the lean blowout limit to
the presence of fuel droplets in the combustion zone. lower equivalence ratios.
Although variable geometry has been used in some large A major drawback of circumferential staging is that "chill-
industrial engines, there have been few successful applications ing" of chemical reactions occurs at the edges of the individual
of this technique in small-to-medium size gas turbines due to combustion zones. This chilling lowers the combustion effi-
size and cost limitations and also because of concerns regarding ciency and increases the formation of CO and UHC. Further-
operational reliability (Aoyama and Mandai, 1984). more, the circumferentially nonuniform exit temperature distri-
bution results in loss of turbine efficiency. These limitations
Staged Combustion. Variable-geometry combustors strive have led to the development of "staged" combustors in which
to maintain the combustion temperature within fairly narrow no attempt is made to achieve all the performance objectives
limits by switching air from one zone to another depending on in a single combustion zone. Instead, two or more zones are
the engine power setting. With staged combustors the air flow employed, each of which is designed specifically to optimize
distribution does not change; instead the fuel flow is switched certain aspects of combustion performance.
from one zone to another in order to maintain a fairly constant The principle of staged combustion is illustrated in Fig. 33.
combustion temperature. One simple method of fuel staging is Increase in overall combustor fuel/air ratio causes the combus-
by "selective fuel injection" as described by Bahr (1987). tion zone temperature to rise. As this temperature approaches
With this technique, which is sometimes called "circumferen- a value above which NO, levels starts to become excessive,
tial staging," fuel is supplied only to selected combinations of fuel is diverted to a separate combustion zone. This approach
fuel injectors at lightoff, relight, and engine idle conditions, as allows the combustor to operate over the entire power range
illustrated in Fig. 32. Only at power settings above idle is the while keeping the combustion zone temperatures within the
full complement of fuel injectors employed. The objective of low-emissions "window." A typical staged combustor has a
this modulation technique is to raise the equivalence ratio and lightly loaded primary zone, which, at low power settings, oper-
hence also the temperature of the localized combustion zones ates at an equivalence ratio of around 0.8 to achieve high com-
at low power operation. This approach, which is now in com- bustion efficiency, low CO and UHC, and also provide the

636 / Vol. 117, OCTOBER 1995 T r a n s a c t i o n s of the ASME

Downloaded From: on 01/27/2016 Terms of Use:

Pilot ~ ~1~ / " / / /

. . . .




OVERALL COMBUSTOR FUEL/AIR RATIO Coo,n0 ~ I U l RH ~ ' ~ . L " - ' ~ -


Fig. 33 Principle of staged combustion

desired temperature rise. At high power settings, it acts as a

pilot source of heat for the main combustion zone downstream
which is supplied with a fully premixed fuel-air mixture. For
full-load operation, both zones are supplied with fuel and the
equivalence ratio in both zones is kept low, around 0.6, to
minimize the emissions of nitric oxide and smoke.
An important choice for the designer is whether the staged
combustion should take place in "series" or in "parallel." The Fig. 35 Fuel nozzle for G,E. dual-annular combustor (courtesy of the
latter approach, often called "radial staging," features the use Parker Hannifin Corporation)
of a dual-annular combustor, as illustrated in Fig. 34. One of
these combustors--usually the outer combustor--is designed
to operate lightly loaded and provides all the temperature rise If the combustor domes of the inner and outer stages are
needed at startup, altitude relight, and engine idle conditions. arranged to be radially in-line, this allows the fuel injector tips
At idle the equivalence ratio of the combustion zone is selected for both stages to be mounted on a common feed arm, as shown
to minimize the emissions of CO and UHC. The other annular in Fig. 34. An important advantage of this arrangement is that
combustor is designed specifically to optimize the combustion the main stage fuel injectors are cooled by the continuously
process at high power settings. It features a small, highly loaded flowing pilot fuel, as illustrated in Fig. 35. This prevents coking
combustion zone of short residence time and low equivalence of the main stage nozzles when they are unfueled but still ex-
ratio to minimize the formation of NOx and smoke. posed to the hot engine environment.
The main advantage of radial staging is that it allows all the There are a number of drawbacks to radial staging. One basic
combustion performance goals to be achieved, including low drawback is that all zones are supplied with air at the compres-
emissions, within roughly the same overall length as a conven- sor outlet temperature, which means that all zones have the
tional combustor. Another advantage of the dual-annular con- same relatively poor lean blowout limit. It is also clear that
cept employed in radial staging is that it provides a satisfactory pollutant reduction is achieved at the expense of increased de-
liner length/height ratio in both annuli within a short overall sign complexity and a marked increase in the number of fuel
length. This short-length feature is attractive from the stand- injectors. The larger liner wall surface area demands additional
points of low engine weight and reduced rotor dynamics prob- cooling air, which has an adverse effect on pattern factor. Fur-
lems (Bahr, 1987). thermore, the peaks of the radial temperature profile could shift
in radial position as a result of fuel staging, with potential
adverse effects on the hot sections downstream of the combus-
tor. Another basic problem with radial staging is that of achiev-
\ ing the desired performance goals at intermediate power settings
where both zones are operating well away from their optimum
design points.
The radially staged combustor shown in Fig. 34 was designed
by the General Electric company. It achieved around 35 percent
reductions in CO and UHC, and 45 percent reduction in NOn
in comparison with the corresponding single-annular combus-
tor. The U.S. Federal Aviation Administration ( F A A ) and the
French Direction Generale de l'Aviation (DGAC) have jointly
certified the CFM56-5B engine with the dual-annular combus-
Preollu=e£tomJ,ztull tot, thereby paving the way for the first flight and certification
Fuel Nozzle ~ l l l ~ l y lngled Liner on Airbus Industrie A320 and A321 aircraft in 1995.
Cooscr~lcttoo With "series" or "axial" fuel staging, a portion of the fuel
Counl:errol:a ¢ttlS is injected into a fairly conventional primary combustion zone.
Additional fuel, usually premixed with air, is injected down-
Fig. 34 General Electric dual-annular combuetor stream into a "secondary" or " m a i n " combustion zone, which

Journal of Engineering for Gas Turbines and P o w e r OCTOBER 1995, Vol. 117 / 637

Downloaded From: on 01/27/2016 Terms of Use:

IGNITORS(2) the water to the outer portion of the air flow path. Another
advantage of having the main zone outside of the pilot is that
MOUNTINGSYSTEM LJ_ / NEW OUTER the radial temperature profile at the combustor outlet peaks
\ ' / _ D,F,U ER OASwAL
toward the outer radius of the turbine flowpath, a situation that
L;,LOT OEL is conducive to long turbine blade life.
In the longer term it is possible that staging of the air flow by
variable geometry in conjunction with fuel staging may become
=~C--"-- 1111i11~ = ~;~ ..... I \ CONVENTIONAL more of a design option.

Dry Low NOr Combustors

The term " d r y " is used to indicate that the combustor in
question is capable of achieving low pollutant emissions, in
particular low NOx, without having to resort to the injection of
• '~
water or steam. Solar Turbines Inc. in San Diego has been
among the pioneers in the development of dry low-emissions
INNER DIFFUSEROASEWALL combustors for industrial gas turbines, and the results of their
efforts have appeared in a number of publications, for example,
Fig. 36 Pratt and Whitney axially staged combustor White et al. (1982), Roberts et al. (1982), Smith et al. (1986,
1991), Smith and Cowell (1989), and Etheridge (1994).
Most of the low-NO~ combustors developed by Solar fall into
operates at low equivalence ratios to minimize the formation the lean premix family of combustors. Figure 37 shows a cross-
of NOx and smoke. The primary combustion zone is used on sectional view of a fuel injector designed for combustors sup-
engine startup and generates the temperature rise needed to raise plied with natural gas. An 18-vane radial flow swirler is used
the rotational speed up to engine idle conditions. At higher to impart a high degree of rotation to the combustor primary
power settings, fuel is supplied to the secondary combustion air, which serves both to promote fuel-air mixing and to induce
zone and, as the engine power rises towards its maximum value, a recirculatory flow in the primary zone. The fuel injector/air
the function of the primary zone becomes increasingly one of swMer assembly permits three different modes of fuel injection,
providing the heat needed to initiate rapid combustion of the as indicated in Fig. 37. Best mixing is achieved by injecting
fuel supplied to the second stage. the gaseous fuel through 18 spokes, each spoke being located
Axial staging does have certain advantages over radial stag- between a pair of swirl vanes. As each spoke contains six holes
ing. Since the main stage is downstream of the pilot stage, of 0.89 mm diameter, the total number of injection points is 108.
ignition of the main stage directly from the pilot is both rapid Combustion tests have been carried out with the fuel injector
and reliable. Also, the hot gas flow from the pilot into the main assembly attached to a cylindrical combustion liner, 200 mm
combustion zone ensures high combustion efficiency from the in diameter and 450 mm long. According to Smith et al. (1986),
main stage, even at low equivalence ratios. According to Segal- these tests showed that the concept is capable of achieving
man et al. (1993), the radial temperature profile at the combus- NOr emissions below 10 ppm when burning natural gas, with
tor exit can be developed to a satisfactory level using conven- simultaneous low values of CO and UHC emissions.
tional dilution hole trimming and, once developed, does not The manner in which the fuel-injection system described
change significantly as a result of fuel staging. above was adapted for liquid fuels by Smith and Cowell (1989)
The main drawback to axial staging is that the in-line arrange- is shown in Fig. 38. The fuel injector/air swirler assembly was
ment of stages tends to create additional length, which makes designed to permit three different modes of liquid fuel injection.
the problem of retrofit difficult for some engines. In comparison The mode designated as "inner filming" in Fig. 38 involves
with conventional combustors, the liner surface area that needs filming of the fuel on the cylindrical swirler centerbody. Fuel
to be cooled is higher. The fuel injectors for the two combustion is delivered to the outer surface of the centerbody through eight
stages require separate feed arms, which involve two separate holes located around the centerbody circumference. This fuel
penetrations of the combustor casing. Furthermore, the pilot is intended to form a film, which is carried downstream to the
fuel cannot be used to cool the main stage fuel as can be done
quite conveniently with radial staging.
Figure 36 shows a cross-sectional view of an axially staged SWIRL
combustor developed by the Pratt and Whitney company (Koff, VANES (18) CHANNEL PILOT
1993). For clarity, the main stage fuel injectors are shown
rotated half an injector pitch to be in line with the pilot stage
injectors. The engine centerline is at the bottom of the figure.
.... ~ITFI~.~'~/iRm'CE¢~/" FUEL

This combustor has the benefits of the axially in-line stage

arrangement without any length penalty and is designed to fit
into the existing P&W B2500-AS engine.
The pilot combustion zone is designed specifically to provide ' ' , I
wide stability limits and high combustion efficiency (low CO
and UHC). With increase in engine power above idle, fuel is
admitted to the main combustion zone where combustion is
initiated and sustained by the hot gas emanating from the pilot
zone. The relative amounts of fuel supplied to the pilot and
main zones is such that no thrust lag is created when fuel is CHANNEL
/ / " - " ~
DOME " ~
first introduced into the main zone. In combination, the pilot INJECTION /
FUEL / |
and main zones maintain a low equivalence ratio, which ensures
low NO~ emissions at higher power settings. INJECTION AIR FLOW
Of special interest in Fig. 36 is the inboard location of the FUEL SPOKE (18)
pilot combustion zone. This greatly reduces the susceptibility
to flame blowout in heavy rain since the compressor centrifuges Fig. 37 Solar Iow-NOx fuel injector for natural gas

638 / Vol. 117, OCTOBER 1995 Transactions of the ASME

Downloaded From: on 01/27/2016 Terms of Use:

Pressure Torch Igniter
Translating Atomizing
Swlrler Air Injectors Riming
Inlet Row ~, Premlxlng J [ Injector
Platk~~:E~ Gas L,._._._..,..i Orifices

C) ~ j t Orifices
I t/ I,/'%.1
ivi Ill
= ~ , Combu=o,
Liner Fer
Methanol / I
Supply Swirler ,-.~, '~
/ I J Swirler
Air L-r-----I--I Centerbody
Row Main o] I oine
Fuel I " -
Fig. 38 Solar Iow-NOx fuel injector for liquid fuels

primary combustion zone by the swirling primary air flow. The

fuel vaporizes and mixes with air as the film progresses along
the centerbody. The inner air assist/airblast air exits through
an annular gap just inside the centerbody lip. Its function is to
break up any remaining fuel film as it flows off the centerbody
and into the primary zone.
The outer filming mode involves film formation on the outer
cylindrical surface of the air swirler channel. Fuel is delivered
to the filming surface in a manner identical to that used for Diffuser Discharge
inner filming. Provision is also made for air blasting the outer nozzle
fuel film at the exit of the swirler channel. Flow visualization Fig. 39 Rolls-Royce Industrial RB 211 DLE 04 module
tests showed that a pressure drop of at least 3 percent is needed
to achieve satisfactory filming of the fuel. At lower pressure
drops, the liquid film is not sustained over the length of the to achieve the required level of mixing between the primary
injector, but tends to form rivulets that swirl to the bottom of efflux and the secondary fuel-air mixture. Combustion testing
the injector and travel as a thick stream to the end of the injector. carried out over a range of pressures from 1 to 20 atm has
Evaluation of the combustor performance demonstrated that demonstrated the ability of this axially staged combustor to
the lowest NO~ emissions were obtained with either total inner achieve simultaneously low NOx and low CO over wide ranges
fueling or combined inner and outer filming. The combustor of power and ambient temperature without resorting to either
yielded around 12 ppm NOx at 6 atm and 20 ppm at 9 atm. CO variable geometry or air bleeds. It was also demonstrated that
was always below 50 ppm. In common with most well-mixed a uniform fuel distribution prior to combustion is essential for
systems, low concentrations of both CO and NOx were attain- achieving low emissions, especially at high pressures. Based on
able only over a fairly narrow range of operating conditions. the test data obtained so far, at full baseload conditions the
Potential improvements for this fuel prefilming combustor predicted engine emissions are 17.4 ppm NOx, 5 ppm CO, and
include increasing the number of fuel injection holes used to zero UHC.
deliver fuel to the filming surface to aid in the formation of a An interesting feature of this combustor is that the level of
more uniform film, and lengthening the injector centerbody to NOx emitted when the primary only is burning is reduced by a
allow a longer time for fuel evaporation and mixing. half when the secondary is ignited, and only when the tempera-
To produce a dry low-emissions combustor for an industrial ture of the secondary zone approaches that of the primary does
version of the Rolls Royce RB 211 engine, the annular aero- the total NOx exceed the primary-only value.
combustor has been replaced by nine radially positioned reverse
flow combustors, as shown in Figs. 39 and 40 (Willis et al.,
1993). This arrangement results in an 80 percent increase in Secondary fuel
combustion volume. The primary zone is fed by two radial air
swirlers. Each swirl passageway has a number of gaseous fuel
injection points. Contrarotating swirlers were selected to mini-
mize the length of the recirculation zone and maximize mixing Central ~ . 12
of fuel and air. The outlet of the primary zone is restricted to dlffuslon~
accelerate the flow and thereby limit the recirculation zone
length and prevent the secondary efflux from becoming en-
trained into the primary zone. The secondary mixing duct is Torch -_~
wrapped around the primary combustor but is separated from Q I Secondary
it by another annular duct, which provides the wall-cooling Prlmary
air. Gaseous fuel is injected into the secondary duct from 36 rnlxlngducts ~
equispaced axial spray bars, each containing six injection holes.

This fuel bar arrangement was determined by trajectory calcula-

tions and an air velocity profile predicted by a CFD code. Fuel Secondary
sampling and combustion tests showed that the resulting fuel- rnlxlng duct
air mixture is uniform to within 4 percent. This mixture is
discharged into the secondary zone at an angle that was selected Fig. 40 Rolls-Royce dry low-emissions premix, lean-burn combustor

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 639

Downloaded From: on 01/27/2016 Terms of Use:


FUEL (a)



Fig. 41 General Electric LM6000 dry Iow-NOx combustor FUEL (b)


Figure 41 gives a cross-sectional view of General Electric's
new LM6000 dry low-emissions combustor, as described by
Leonard and Stegmaier (1994). This premix combustor requires
about twice the volume of the conventional annular combustor
it replaces in order to reduce the emissions of CO and UHC.
Part of the air used in combustion, which at maximum power
is around 80 percent of the total combustor airflow, flows into
the combustion zone through three annular rings of premixers, INNER ~ MIXING
as shown in Fig. 41. The two outer rings each have 30 fuel- ...... DUCT
air premixers while the inner ring has 15. This arrangement of
premixers facilitates fuel staging at part-load operation. The (c)
total of 75 fuel nozzles is formed by having 15 stems with 3
premixers on each stem, as shown in Fig. 42. Each stem incorpo- CENTERBODY
rates three separate fuel circuits for independently fueling the
three premixers.
A short annular liner was selected to minimize the amount
of air needed for wall cooling. Only backside cooling is used,
so a thermal barrier coating is applied to both the liner and in Fig. 43 Cross-sectional views of three mixer designs

the dome area to keep the metal temperatures within acceptable

limits. The use of a multipass diffuser also permits further re-
duction in overall combustor length. Of special interest and
importance to the attainment of low emissions is the design of
the premixers. Figure 43 shows cross-sectional views of three
different mixer designs that were subjected to combustion test-
ing. The Double Annular Counter Rotating Swirler (DACRS)
was conceived to satisfy the restraints of autoignition, flashback
and size. Its exit velocity is determined by the need to greatly
exceed the turbulent flame speed of gaseous fuel-air mixtures.
The duct diameter is reduced at the exit in order to create an
accelerating flow and thereby prevent flashback. The conical
centerbody located along the centerline of the premixer can be
used to supply liquid fuel to an atomizer at its tip, and gas
passages for diffusion burning at low power conditions (Joshi
et al., 1994).
The design objective with this type of mixing device is to
produce a completely homogeneous mixture of fuel and air at
the premixer exit. As the total area of the fuel injection holes
is fixed by the flow rate and the available fuel injection pressure,
the design procedure is essentially one of finding the best com-
promise between the desire for a large number of fuel injection
points and the equally important requirement of large injection
holes to allow the fuel jets to penetrate across the air stream.
In the premixer design designated as DACRS I in Fig. 43 (a),
the fuel is injected into the air stream radially outward from
holes in the centerbody just downstream of the swirl vanes.
This configuration gave low NOx at some test conditions but
Fig. 42 Fuel nozzle assembly for LM6000 dry Iow-NOx combustor not at others. This was attributed to unsatisfactory fuel jet pene-

640 / Vol. 117, OCTOBER 1995 Transactions of the A S M E

Downloaded From: on 01/27/2016 Terms of Use:

tration, so a modification was made by adding eight radial
spokes in the location of the holes in the centerbody, as shown
in Fig. 43 (b). Each spoke has three holes to inject gaseous fuel
~I lance
perpendicular to the flowing air stream. Combustion testing of
this DACRS II mixer showed that single-digit NOx emissions
are attainable with this concept. A further modification to the
premixer designs described above was made by incorporating
fuel injection holes into the swirl vanes of the outer swirler, as
illustrated in Fig. 43(c). In this DACRS III configuration, the
fuel is injected through three holes in the trailing edge of each
outer swirl vane and one hole in the outer wall of the mixing
duct in between each swirl vane. Fuel is fed to the hollow outer
vanes through a manifold on the outside of the premixing duct.
The NOx emissions obtained with the DACRS III mixer were
very similar to the DACRS II design.
A big advantage of this type of premixer module is that,
once developed, it has broad applications in a wide range of
combustor sizes. The basic module remains the same regardless
of combustor size; only the number varies. According to Joshi
et al., the DACRS II and DACRS III mixers could be applied
to a range of GE engines, including the LM1600, LM2500, and
LM6000, since single-digit NOx emissions have been attained
with both these mixers at test conditions encompassing the op-
erating regions of these engines.
The ABB company has developed a Conical Premix Burner
module, called the EV-burner, which also appears to have good to turbine from compressor
potential for a wide range of dry low-emissions combustion
applications. A cross-sectional view to illustrate the operating Fig. 45 Silo combustor equipped with conical premix burner
principles of the burner is given in Fig. 44 from Sattelmayer et
al. (1992). A unique feature of this burner is the flame stabiliza-
tion in free space near the burner outlet, which utilizes the
sudden breakdown of a swirling flow. The swirler consists of the baseload conditions of 12.5 bar and Ti,let = 643 K, NOx
two halves of a cone, which are shifted to form two air inlet values of 13 ppm were measured (Aigner and Muller, 1993).
slots of constant width. The device is claimed to be equally The EV-burner technology has also been applied to the design
satisfactory for both gaseous and liquid fuels. Gaseous fuels are of an annular combustor for the heavy duty ABB GT13E2 gas
injected into the combustion air by means of two fuel distribu- turbine (Senior et al., 1993). This is a single shaft industrial
tion tubes comprising two rows of small holes, which are lo- engine in the large size range ( > 150 MW). The EV burners
cated parallel to the inlet parts of the swirler. Fairly complete have a double offset row arrangement within the annular com-
mixing of fuel and air is obtained shortly after injection. Liquid bustor. As the power setting is reduced, the overall equivalence
fuels are injected at the apex of the cone using a pressure or ratio eventually becomes too low to allow lean premix operation
air-assist type of atomizer. Fuel evaporation and mixing of fuel in all burners. The technique used to overcome this problem is
vapor with air is accomplished within the burner and upstream to maintain some burners at a stable equivalence ratio while
of the recirculation zone. In contrast to most LPP designs, no others run leaner but are stabilized by the first burner group.
diffusion or pilot stage is needed to improve the stability of the Combustion tests carried out at atmospheric pressure on a
premix flame. Also, because the combustion zone is displaced scaled-down combustor have shown that the presence of these
from the burner walls, the heat transfer to the burner section is "cold" burners has no significant effect on combustor pattern
minimized. factor. These tests also indicated very low values of CO, UHC,
In February 1991 an ABB GT11N gas turbine was retrofitted and NOx, in agreement with previous experience of EV burners.
with a new silo combustor of the type shown in Fig. 45. This The strict NOx regulations in Japan have promoted several
silo combustor is equipped with 37 EV burners, all of which recent developments in dry low-NOx combustion. During the
operate in a pure premix mode. For part-load operation, fuel is past 5 years, Kawasaki Heavy Industries Ltd., a manufacturer
supplied to only a fraction of the total number of burners At of small-to-medium size gas turbines, has been developing a
dry low-NOx combustor burning natural gas for application to
a 1.5 MW gas turbine. The tubular combustor employs a combi-
nation of lean premixed combustion and fuel staging (multiple
fuel injectors). The pilot nozzle is a conventional multi-orifice
Dilution Al r . ~ type, which supplies up to 5 percent of the total fuel. The eight
Combustion Air ~-A.-------'~-",-. w ,\ main nozzles are subdivided into four groups, with two in each
group. At full load all eight nozzles are flowing. With gradual
reduction in load, successive groups of nozzles are shut off,
until at idle only the pilot nozzle and one group of main nozzles
v~; are supplied with fuel. This combustor has demonstrated a maxi-
Fu., ,~ Spray Evaporation ~ .*..*~ mum NOx level of 42 ppm and work is in progress to reduce
this to 20 ppm (Kajita et al., 1993).
Nakata et al. (1992) have designed an RQL combustor to
burn a coal-gasified fuel having a calorific value of only one
Atomization ~ " " ~ ' - "~'~ tenth that of natural gas. Its main combustible component is
(Swirl Nozzle) Gas Injection ~ ~ ~.,.. Flame
Holes ~ Front carbon monoxide, but it also contains appreciable concentra-
tions of ammonia, which is converted into NOx during the com-
Fig. 44 Operating principle of ABB conical premix burner bustion process. Even so, this combustor achieved an NOx level

Journal of Engineering for Gas Turbines and P o w e r O C T O B E R 1995, Vol. 117 / 641

Downloaded From: on 01/27/2016 Terms of Use:

and fuel-air mixing. Its function is to achieve complete evapo-
ration and complete mixing of fuel and air prior to combustion.
By eliminating droplet combustion and supplying the combus-
tion zone with a homogeneous mixture of low equivalence ratio,
the combustion process proceeds at a uniformly low temperature
and very little NOx is formed. In the second region the flame
is stabilized by the creation of one or more recirculation zones.
Combustion is completed in this region and the resulting prod-
ucts then flow into region three, which may comprise a fairly
conventional dilution zone.
A useful byproduct of LPP combustion is that it is essentially
free from carbon formation, especially when gaseous fuels are
Fig. 46 Schematic cross-section of venturi mixer for the ALEC combus- used, in which case the description "lean premixed combustor"
tor is more appropriate. The absence of carbon not only eliminates
soot emissions but also greatly reduces the amount of heat
transferred to the liner walls by radiation, thereby reducing the
of 60 ppm when burning the standard fuel containing 1000 ppm amount of air needed for liner wall cooling. This is an important
of ammonia. consideration because it means that more air is made available
The Japan Automobile Research Institute is collaborating for lowering the temperature of the combustion zone and im-
with the Toyota Central Research and Development Laboratory proving the combustor pattern factor.
in the development of a low-emissions combustor for a 100 Another important advantage of LPP systems is that for
kW automotive ceramic gas turbine (Kumakura et al., 1994). flames in which the temperature nowhere exceeds 1900 K the
Studies are being carried out in a premix-prevaporize tube op- amount of NOx formed does not increase with increase in resi-
erating both at atmospheric pressure and on the engine at high dence time (Anderson, 1975; Leonard and Stegmaier, 1994).
pressures to provide quantitative data on the influence of fuel This means that LPP systems can be designed with long resi-
drop size and fuel distribution on the degree of vaporization dence times to achieve low CO and UHC, while maintaining
achieved (Ohkubo et al., 1994). low NOx levels. This finding is especially significant for indus-
Other dry low-NOx combustor concepts for small gas turbines trial engines, where size is less important than for aero-engines.
include the Annular Low-Emission Combustor (ALEC), which According to Leonard and Stegmaier, this approach leads to an
is being developed in several forms by Gasunie Research in the LPP combustor volume that is approximately twice that of a
Netherlands (Bahlmann and Visser, 1994). It is basically a conventional combustor.
lean-premix combustor in which mixing of gaseous fuel and air The main drawback to the LPP concept is that the long time
is accomplished using a venturi premixer, as illustrated in Fig. required for fuel evaporation and fuel-air premixing upstream
46. The venturi extends inwardly (tangentially) into the annular of the combustion zone may result in the occurrence of autoigni-
combustion liner and is double-walled and air-cooled. To avoid tion or flashback at the high inlet air temperatures associated
combustion within the venturi mixers, the diameter of the ven- with operation at high power settings. Another drawback is that
turi outlet is such that the average flow velocity at exit is 30 some form of piloting device may be needed to facilitate ignition
m/s. The diffuser angle of the venturi is set at 5.5 deg. Prelimi- and sustain combustion at arduous operating conditions. Also,
nary tests with a 7 deg angled venturi resulted in flow separation the airflow rate needed for lean combustion during high-power
and combustion inside the venturi. The ALEC concept has dem- operation may result in flame blowout at low power conditions.
onstrated very low NOx levels ( < 10 ppm) along with satisfac- To alleviate this problem, which is endemic to the LPP combus-
tory CO levels ( < 5 0 ppm). tor due to its need to operate close to the lean blowout limit,
some form of fuel staging or variable geometry is usually re-
Lean Premix Prevaporize Combustion Another problem associated with all well-mixed combustion
As the turbine inlet temperatures of many gas turbines are systems is that of acoustic resonance, which occurs when the
well below the temperatures needed to produce NOx in signifi- combustion process becomes coupled with the acoustics of the
cant quantities, it should, in principle, be possible to achieve combustor. This phenomenon has been encountered many times
low levels of NOx without any sacrifice of cycle efficiency. In in the past and is often manifested merely as noise, but in severe
practice, the requirements of flame stability and high heat re- cases it creates mechanical vibrations, which may cause damage
lease rates usually demand flame zones in which the gas temper- to engine components. Improvements in fuel-air mixing always
ature is appreciably higher than the turbine inlet temperature. encourage the onset of combustion noise. This problem is there-
Conventional combustors typically inject fuel directly into the fore of great importance to the future development of lean pre-
primary combustion zone with little or no premixing with air. mixed combustors.
Consequently, combustion takes place in local regions at near- In summary, lean premix prevaporize combustion has consid-
stoichiometric temperatures with consequent high rates of NOx erable potential for ultralow-NOx emissions. Several workers
formation. have reported NOn levels below 10 ppm, even with flame tem-
A common feature of all the dry low-NOx combustors de- peratures higher than 2000 K (Poeschl et al., 1994). However,
scribed above is that positive efforts are made to eliminate these many formidable problems remain, the principal being that of
local regions of high temperature by mixing the fuel and air achieving complete evaporation of the fuel and thorough mixing
upstream of the combustion zone. The lean, premix, prevaporize of fuel and air within the autoignition delay time and without
(LPP) concept represents the ultimate in this regard. Its underly- risk of acoustic resonance or flashback.
ing principle is to supply the combustion zone with a completely
homogeneous mixture of fuel and air, and then to operate the Droplet and Spray Evaporation. The evaporation of
combustion zone at an equivalence ratio which is very close to drops in a spray involves simultaneous heat and mass transfer
the lean blowout limit. The smaller the margin between stable processes in which the heat for evaporation is transferred to the
combustion and flame blowout, the lower will be the output drop surface by conduction and convection from the sur-
of NOx. rounding air or gas, and vapor is transferred by convection and
A typical LPP combustor can be divided into three main diffusion back into the gas stream. The overall rate of evapora-
regions. The first region is for fuel injection, fuel vaporization, tion depends on the pressure, temperature, and transport proper-

6 4 2 / Vol. 117, O C T O B E R 1995 T r a n s a c t i o n s of th e A S M E

Downloaded From: on 01/27/2016 Terms of Use:

ties of the gas; the temperature, volatility, and diameter of the 1.6 I I I I I I I
drops in the spray; and the velocity of the drops relative to that P= IO0 kPa
of the surrounding gas.
If a single-component fuel drop is suddenly immersed in gas
at high temperature, initially almost all of the heat supplied to
the drop serves to raise its temperature. As the fuel temperature

~m/UD°' S)(/.Lm]
rises, fuel vapor is formed at the drop surface and part of the
heat transferred to the drop is used to furnish the heat of vapor-
~ 5000
ization of the fuel. Eventually, a stage is reached where all of 1.0
the heat transferred to the drop is used as heat of vaporization
and the drop stabilizes at its "wet-bulb" or "steady-state" 0.8
temperature. Thus the total drop evaporation time can be subdi-
vided into two main components, one for the heat-up period 0.6 Too= 2000 K o
and another for the steady-state phase.
During the steady-state phase, evaporation rates are relatively "'
high, but during the initial heat-up period much of the heat 1.0
transferred to the drop is absorbed in heating it up, so the amount
of heat available for fuel vaporization is correspondingly less. 0.8
This lower rate of vaporization, when considered in conj unction =
with the significant proportion of the total drop lifetime occu- N"
pied by the heat-up period, especially at high pressures, means ~ 0.6
that overall evaporation rates can be appreciably lower than the ~.
experimental values quoted in the literature, most of which were '~ 0.4
measured during steady-state evaporation at normal atmo- 200
spheric pressure, i Too = 12OO K =o~
In most continuous flow combustors, the fuel is sprayed into 0.2
air or gas flowing at high velocity. Where relative motion exists
between the droplets and the surrounding gas, the rate of evapo- 0.24
ration is enhanced. The effect of convection on evaporation
rates can be accommodated by applying a correction factor to
the evaporation rate calculated for quiescent conditions. Where 0.213
heat transfer rates are controlling, the correction factor is
1 + 0.22 Re~ 5
where ReD, the drop Reynolds number, is typically around 5.
From a practical viewpoint it would be very convenient if
the effect of the heat-up period could be combined with that of O.Oe
forced convection in a manner that would allow an "effective"
value of evaporation constant to be assigned to any given fuel 0
0.04 I ! I I I I I
at any stipulated conditions of ambient pressure, temperature, 420 440 460 480 500 520 540 56O
velocity, and drop size. To accomplish this, Chin and Lefebvre Tbn, K
(1982) defined an effective evaporation constant as
Fig. 47 Variation of effective evaporation constant with normal boiling
point for a pressure of 100 kPa
horf = D2/te (9)

where te is the total time required to evaporate the fuel drop,

including both convective and transient heat-up effects, and Do while the average rate of fuel evaporation is readily determined
is the initial drop diameter. as
Calculated values Of ~ff for an ambient air pressure of 100
kPa are plotted in Fig. 47. Similar plots for higher levels of my = ( T r / 6 ) p F ~ f f D o (11)
pressure may be found in Chin and Lefebvre (1982). Figure
The velocity term U in Fig. 47 is the relative velocity between
47 shows plots of Xofeversus Tbn, the normal boiling point, for
the drop and the surrounding gas. However, small droplets rap-
various values of UDo at three levels of ambient temperature,
idly attain the same velocity as the surrounding gas, after which
namely, 500, 1200, and 2000 K. While recognizing that no
they are susceptible only to the fluctuating component of veloc-
single fuel property can fully describe the evaporation character-
ity, u'. Thus for gas turbine combustors, where the value of u '
istics of any given fuel, the normal boiling point, Tbn, has much
is usually high enough to affect evaporation rates, U in Fig. 47
to commend it for this purpose, because it is directly related to
should be replaced with u'.
fuel volatility and vapor pressure. ]t also has the virtue of being
For sprays, the average rate of fuel evaporation is obtained
quoted in fuel specifications. Figure 47 shows that hereincreases
with increases in ambient temperature, pressure, velocity, and
drop size and diminishes with increase in normal boiling tem- mF = pm~.effVq/D2o (12)
The concept of an effective value of evaporation constant where PA is the air or gas density, V is the spray volume, q is
considerably simplifies calculations of the evaporation charac- the fuel/air ratio by mass, and Do is the Sauter mean diameter
teristics of fuel drops. For example, for any given conditions of the droplets in the spray.
of pressure, temperature, and relative velocity, the lifetime of Equation (12) may be used to calculate average rates of
a fuel drop of any given size is obtained from Eq. (9) as fuel spray evaporation, while Eq. (10) gives the length of duct
required for complete evaporation if the spray is injected into
te = Do2/~ff (10) a ducted air stream. The drop diameter selected for insertion

Journal of Engineering for Gas T u r b i n e s and P o w e r O C T O B E R 1995, Vol. 117 / 643

Downloaded From: on 01/27/2016 Terms of Use:

10 for a nearly perfect premixer, a well-designed prernixer, and
a nonoptimized premixer. It clearly illustrates the tremendous
advantage to be gained from thorough mixing of air and fuel.
These workers also noted that the amount of NOx formed in a
5 z ++ nonoptimized premixer increased with increasing pressure.
z They attributed this result to the fact that reactions taking place
in the hot spots (>2000 K) of poorly premixed flames are
LU pressure dependent.
§2. PREMiXE~..Ii-.F
The only exception to the general rule that better premixing
yields less NOx are the results obtained by Santavicca et al.

S l I I I I
(1993). These workers examined the effects of incomplete
fuel-air mixing on the lean blowout limits and emissions char-
1650 1700 1750 1800 1850 1900 1950 acteristics of an LPP coaxial mixing tube combustor. A fluores-
FLAME TEMPERATURE, K cence technique was used to determine the degree of fuel vapor-
ization and mixing at the combustor inlet. Contrary to what was
Fig. 48 Effects of nonuniform fuel-air mixing on NO. formation expected, it was found that an improvement in fuel-air mixing
resulted in comparable NOx emissions for the same conditions
of inlet temperature and equivalence ratio. Although the better
into Eq. (10) should, of course, be that of the largest drop in mixed device demonstrated lower NOx emissions, this was at-
the spray. tributed to its ability to operate at lower equivalence ratios.
In a number of recent publications, Chin (1994, 1995a, b) The formidable problems involved in trying to attain com-
has developed and extended the h~ff concept to include several plete homogeneity in the fuel-air mixture entering the combus-
phenomena of practical importance, including the secondary tion zone have prompted studies into the use of mechanical
breakup of droplets as well as real gases, near-critical tempera- mixers as a means of achieving the desired degree of fuel-air
ture, and multicomponent fuel effects. Recent advances in the mixing. Static mixers are well known and widely used in pro-
modeling of droplet vaporization have been reviewed by Peng cess engineering for mixing of both gases and liquids, but as
and Aggarwal (1995). Their review also includes the methodol- yet they appear to have evoked little interest for combustion
ogies currently available for representing droplet motion and applications. Poeschl et al. (1994) examined the mixing capabil-
vaporization history in two-phase flow computations. ity of a commercially available static mixer after a series of
tests using pressure, airblast, air-assist, and multipoint injectors
Mixing. The attainment of perfect mixture homogeneity had failed to provide the right degree of homogeneity. All these
prior to combustion is of paramount importance to the success injection devices had either failed to mix fuel vapor with the
of most types of low-emissions combustors. The influence of air flowing near the duct wall or had deposited liquid fuel on
mixture inhomogeneity on NOx formation has been examined the wall. Tests with the static mixer yielded excellent homoge-
by several workers, both theoretically and experimentally. Ap- neity, with a standard deviation of lower than 5 percent for a
pleton and Heywood (1973) employed a burner in which kero- 2 percent pressure loss. This high degree of homogeneity in the
sine fuel was injected using an air-assist atomizer, which al- combustible mixture is very desirable, not only from an emis-
lowed the degree of fuel-air mixing to be varied by changing sions standpoint, but also because it greatly reduces the possibil-
the atomizing air pressure. For low equivalence ratios, they ity of spontaneous ignition. Fuel-lean mixtures tend to have
found that NOx emissions decreased by a factor of ten as a long autoignition delays, but if imperfections in mixing result
result of better fuel-air mixing. In a similar study by Semerjian in local regions in which the equivalence ratio is higher than
et al. (1979) it was found that poorer fuel-air mixing resulted the average value, the ignition delay time could be greatly re-
in increased NOx emissions for overall fuel-lean flames. Lyons duced. Thus a high degree of mixture homogeneity is essential,
(1981) used a multipoint fuel injector spraying Jet A fuel to not only for the attainment of low NOx emissions, but also to
achieve different equivalence ratio profiles across the diameter alleviate the problems of autoignition and flashback. Flashback
of the flametube. The results showed that spatial nonuniformity is unlikely to occur with static mixers if the velocity through
in equivalence ratio resulted in increased NOx emissions for the mixer is kept high ( >20 m/s). Also, by flattening the veloc-
equivalence ratios below 0.7 and decreased NOx emissions for ity profile, the mixer eliminates the boundary layer along which
near-stoichiometric mixtures. flashback is most prone to occur.
Flanagan et al. (1992) used a simple mixing tube fitted with According to Valk (1994) the static mixer system has good
a bluff-body flameholder at its exit. By changing the location potential for engines of modest compression ratio, but more
of the natural gas fuel injector along the length of the tube, the work is needed to reduce length, residence time, and pressure
degree of fuel-air mixing in the mixture approaching the stabi- loss, while maintaining good mixing performance.
lizer could be varied. When the system was operating at an
equivalence ratio of 0.66, a nearly fivefold increase in NO+ Spontaneous Ignition. Spontaneous ignition, or autoigni-
emissions was recorded when going from well-mixed to incom- tion, is a process whereby a combustible mixture undergoes
pletely mixed conditions. chemical reaction, leading to the rapid evolution of heat in the
Fric (1992) used an experimental apparatus very similar to absence of any concentrated source of ignition such as a flame
that employed by Flanagan et al. to examine the NOx emissions or spark. In the LPP combustor and other types of low-emission
produced when burning natural gas fuels at normal atmospheric combustors in which fuel and air are premixed prior to combus-
pressure. He found that temporal fluctuations in equivalence tion, spontaneous ignition must be avoided at all costs since it
ratio can also raise NOx emissions, in addition to spatial nonuni- could damage combustor components and produce unacceptably
formities. For example, temporal fluctuations of 10 percent re- high levels of pollutant emissions.
suited in a doubling of NOn emissions. Spontaneous ignition delay may be defined as the time inter-
Leonard and Stegmaier (1994) used a gas-fired GE LM6000 val between the creation of a combustible mixture, say by in-
combustor to examine the effects of premixing on NO+ forma- jecting fuel into a flowing air stream at high temperature, and
tion. The results obtained are given in Fig. 48, which shows the onset of flame. In view of their practical importance, mea-
NO+ as a function of average flame temperature for various surements of spontaneous ignition delay time have been con-
degrees of premixing. Nonuniformities are the result of fluctua- ducted for many fuels over wide ranges of ambient conditions
tions in time as well as variations in space. Figure 48 gives data and in a variety of test vehicles, including constant-volume

644 / Vol. 117, OCTOBER 1995 Transactions of the ASME

Downloaded From: on 01/27/2016 Terms of Use:

bombs, rapid-compression machines, shock tubes, and continu- 1
ous-flow devices. The test methods employed and the results
obtained are described in reviews by Mullins (1955), Spadac-
cini and Te Velde (1980) and, more recently, by Goodger and
Eissa (1987) and Lundberg (1994).
Ignition delay times are frequently correlated using empirical
equations of the form
~- = A . P - " . e x p ( B / T ) 13)
rn 10"1
T0, K
where the values of A and B are determined by experiment. A is
a proportionality constant that has dimensions, B is an effective F-
activation energy, P is the pressure, and T is a characteristic 600
temperature, usually the initial temperature of the gas (or vapor) a
fuel-air mixture. z
Wolfer (1938) first proposed this form of equation. His anal-
ysis of experimental data showed that
~ 10.2
"7- = 0.43P -H9 exp(4650/T) (14)
where 7- is the ignition delay time in ms, P is the pressure in
atm, and T is the initial mixture temperature in degrees Kelvin.
Despite its longevity, Eq. (14) is still in widespread use. To
accommodate the effects of equivalence ratio on ignition delay r ~ 800
times, it is sometimes modified to 100o
T = AP-"qh .... e x p ( B / T ) (15)
10"3102 102 104
There is little agreement between different workers in regard
to the influence of equivalence ratio on ignition delay time. For PRESSURE, kPa
methane, propane, and aviation kerosine, Lefebvre et al. (1986)
Fig. 49 Variationof ignitiondelaytime with pressureand temperature;
reported values for m of 0.19, 0.30, and 0.37, respectively. SMD = 50 ~um,u' = 0.25 m/s
Mullins (1955) observed no effect of equivalence ratio on igni-
tion delay time, while Ducourneau (1974) and Spadaccini and
Te Velde (1980) both found strong effects. The explanation for Most analyses and equations for ignition delay time ignore
these differences probably lies in the mode of fuel injection. the effects of fuel vaporization, which is reasonable under con-
As discussed above, with liquid fuels there is always the poten- ditions where the evaporation time is appreciably shorter than
tial for stoichiometric combustion in regions close to the evapo- the mixing and reaction times. However, it is important to bear
rating spray. Thus, the ignition delay times will always tend to in mind that the spontaneous ignition delay time results from
correspond to those for stoichiometric mixtures, even though both physical and chemical processes. For liquid fuels, the phys-
the average equivalence ratio of the mixture may be appreciably ical delay is the time required to heat and vaporize the fuel
below the stoichiometric value. Just how close will depend on drops and to mix the fuel vapor in flammable proportions with
the mean drop size of the spray, since this governs the rate of the surrounding air. The chemical delay is the time interval
fuel evaporation and hence also the length of time that stoichio- between the formation of a flammable mixture and the appear-
metric "streaks" of fuel-air mixture can survive, and on the ance of flame. Thus, the physical processes dominate in the
number of fuel injection points. In this context it is of interest early stages of spontaneous ignition, while in the later stages
to note that Tacina ( 1983 ) obtained much more consistent auto- the chemical processes become overriding.
ignition data with a single orifice injector than with a 41-hole Rao and Lefebvre ( 1981 ) have proposed a model for sponta-
injector, which ostensibly provided a more uniform fuel-air neous ignition that takes full account of both chemical and
mixture. Presumably this was because with a single injector the physical effects and has general application to both homoge-
rate of fuel-air mixing was so slow that the bulk of the prereac- neous and heterogeneous mixtures, including situations in
tions leading up to the onset of ignition took place in near- which both fuel drops and fuel vapor are present at the outset
stoichiometric mixtures regardless of the average equivalence of the preignition process. The spontaneous-ignition delay time
ratio. With gaseous fuels, the inconsistencies associated with t is obtained as the sum of the times required for evaporation
slow fuel evaporation are no longer present, but the measured and chemical reaction
ignition delay times are still very dependent on the time required
for the fuel and air to combine in combustible proportions. As t = (D~/k~ff)[1 - (1 _ f)o.67]
with liquid fuels, the longer the mixing time the closer will + 4.66.10 -8 e x p ( 9 1 6 0 / T , , ) / p A f 4 9 (16)
the measured ignition delay times approach the stoichiometric
values. where ~b is the equivalence ratio, T,,,is the initial mixture temper-
In summary, autoignition data are very apparatus dependent ature, and f i s the fraction of the total fuel that must be converted
and, in particular, very fuel-injector dependent. Considerable to vapor to initiate the chemical reaction. As little or no chemical
caution should be exercised in comparing and selecting autoig- reaction takes place at equivalence ratios much below 0.5, a
nition data and in no circumstances should experimentally de- value of f i s recommended for insertion into Eq. (16) such that
rived equations be extrapolated to pressures and temperatures the product of f and q5 is 0.5.
outside the range of experimental verification. This is because Calculated values of t from Eq. (16) are shown plotted
differences in reaction routes may occur over different levels against pressure in Fig. 49 for three levels of temperature,
of temperature and pressure. Only if the available experimental namely 600, 800, and 1000 K. Typical values for u' and Do of
data or calculated values of ignition delay time exceed the resi- 0.25 m/s and 50 #m respectively, were used in calculating heft'.
dence time by a factor of at least five can the designer feel The results show that the influence of fuel evaporation on t is
confident that spontaneous ignition is unlikely to present prob- quite small at the lowest temperature considered (600 K) but
lems. becomes increasingly significant with increasing temperature.

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 645

Downloaded From: on 01/27/2016 Terms of Use:

\ discourages NOx formation from fuel-bound nitrogen (M~tin
and Dederick, 1976). For RQL combustion to be fully effective,
the fuel must be finely atomized and uniformly distributed
throughout the rich combustion zone. Moreover, the primary-
zone airflow pattern must be designed to prevent the occurrence
of localized flow recirculation zones which can increase resi-
dence times and thereby increase NOx production (Micklow et
Fig. 50 Schematic diagram of Rich-burn, Quick-quench, Lean-burn al., 1993).
(RQL) combustor As the fuel-rich combustion products flow out of the primary
zone, they encounter jets of air that rapidly reduce their tempera-
ture to a level at which NOx formation is negligibly small.
This explains the reduced dependence of t on P with increase Transition from a rich zone to a lean zone must take place
in temperature, as shown in Fig. 49. rapidly to prevent the formation of near-stoichiometric, high
The model proposed by R a t and Lefebvre (1981) shows that NOx-forming streaks. The ability to achieve near-instantaneous
fuel evaporation times are negligibly small in comparison with mixing in this "quick quench" region is the key to the success
chemical reaction times for well atomized, highly volatile fuels, of the RQL concept.
especially at conditions of low pressure and temperature. How- An important design consideration is the temperature of the
ever, it also shows that fuel evaporation times become increas- lean-burn zone. If this temperature is too high, the production
ingly significant with increase in pressure and temperature, and of thermal NOx becomes excessive. On the other hand, the
could be rate-limiting for the autoignition of some of the alterna- temperature must be high enough to eliminate any remaining
tive fuels that are now being considered, many of which have CO, UHC, and soot. Thus, the equivalence ratio for the lean-
low evaporation rates due to their high viscosity (poor atomiza- bum zone must be carefully selected to satisfy all emissions
tion) and low volatility. requirements. Typically, lean-burn combustion occurs at equiv-
alence ratios between 0.5 and 0.7 (Talpallikar et al., 1992).
Flashback. An intrinsic feature of all premixed-fuel com- After the requirements of combustion and liner wall cooling
bustion systems is a tendency toward flashback. Flashback oc- have been satisfied, any remaining air can be used as dilution
curs when the flame travels upstream from the combustion zone air to tailor the exit temperature pattern for maximum turbine
into the premixing sections of the combustor. This upstream durability.
propagation of flame takes place whenever the flame speed In some designs the atomizing air is arranged to flow over
exceeds the approach flow velocity. the outside of the liner wall in the rich zone before entering the
Two main types of flashback have been identified: ( 1 ) flash- fuel nozzle. This regenerative backside convective cooling is
back occurring in the free stream and (2) flashback occurring an important design feature because conventional film cooling
through the low-velocity flow in the boundary layer along the in the rich zone would create local near-stoichiometfc mixtures
walls of the premixing section. Either mechanism may involve that would produce high levels of NOx.
homogeneous and/or heterogeneous reactions. Most of the work carried out so far on the RQL concept has
The most obvious free-stream mechanism would involve confirmed its potential for ultralow-NOx combustion. It also has
flashback due to a flow reversal in the bulk flow through the an important practical asset in that it requires only one stage of
combustor. This flow reversal could be a result of compressor fuel injection. However, in order to exploit this potential fully,
surge or combustion instability. Flashback can also occur in the significant improvements in the quench mixer design are
absence of flow reversal if the turbulent flame speed through needed. Other inherent problems include high soot formation
the gas in the premixing section is greater than the local bulk in the rich primary zone, which gives rise to problems of high
velocity. Lean combustion tends to reduce flame speeds, but flame radiation and exhaust smoke. These problems are exacer-
other factors associated with the engine cycle, such as high bated by long residence times, unstable recirculation patterns,
temperatures, pressures, and turbulence levels, and preignition and nonuniform mixing.
reactions in the gas due to appreciable residence times at high Early experimental studies by Novick et al. (1982) suggested
temperature levels, cause increased flame speed. Therefore, that NOx emissions appear to be controlled only by inlet temper-
flame speeds may be sufficiently high to necessitate high flow ature and rich zone equivalence ratio. Changes in the lean or
velocities in the premix-prevaporize section. dilution zone had no observable effect on NOx. On the other
The boundary-layer mechanism involves flashback through hand, carbon monoxide and smoke emissions were influenced
retarded flow in a boundary layer. Important relevant parameters markedly by both rich-zone and lean-zone equivalence ratios,
include the wall temperature and temperature distribution, and as well as by combustor inlet temperature. A minimum lean-
the boundary-layer structure, turbulence properties, and thick- zone equivalence ratio of 0.6 was needed to achieve satisfactory
ness. For more detailed information on flashback, reference smoke levels. In more recent studies, Rizk and Mongia (1991a,
should be made to Plee and Mellor (1978). 1993a, b) have applied three-dimensional emissions modeling,
using well-established reaction mechanisms to RQL combus-
tion. Their results generally confirm the previous findings of
Rich-Burn, Quick-Quench, Lean-Burn Combustor Novick et al. in regard to the importance of rich-zone equiva-
One combustor concept that has been identified as a leading lence ratio to NOx emissions, but they also stress the contribu-
candidate for low NOx emissions is the Rich-burn/Quick- tion to NOx formation of residence time and combustion pres-
quench/Lean-burn (RQL) combustor. Originally conceived for sure.
industrial engines (Mosier and Pierce, 1980; Pierce et al., At the present time the RQL concept is being actively studied
1980), the RQL concept utilizes staged burning, as shown in by the Pratt and Whitney Company and other laboratories as
Fig. 50. Combustion is initiated in a fuel-rich primary zone at part of NASA's HSCT (High Speed Civil Transport) program.
equivalence ratios of between 1.2 and 1.6, thereby reducing The aim of this program is to demonstrate the feasibility of
NOx formation by lowering both the flame temperature and the attaining NOx levels of 3 to 8 g/kg fuel (approx. 40 to 100
available oxygen. Higher equivalence ratios would be even ppm) at supersonic cruise conditions with kerosine fuel.
more beneficial from a NOr standpoint, but could lead to exces-
sive soot formation and smoke. The hydrocarbon reactions pro- Catalytic Combustors
ceed rapidly, causing further depletion of oxygen and further The very strong dependence of NOx formation on flame tem-
inhibiting NOx formation. This initial fuel-rich combustion also perature means that NOn emissions are lowest when the combus-

646 / V o l 117, O C T O B E R 1995 T r a n s a c t i o n s of the A S M E

Downloaded From: on 01/27/2016 Terms of Use:

Preheat MultI-Polnt/Venturl Catal~t ]Bed Post-Catalyst
Combustor Premlxer Combustar

Fart L o a d
Fuel Injection

Fig. 52 Subscale catalytic combustor for industrial applications

Fuel-air Catalytic Thermal Dilution

preparation reactor reactor
subsequently oxidized in the catalyst. The preheat burner is a
conventional well-mixed combustor, which is used to preheat
Fig. 51 Schematic representation of catalytic combustor the premixed fuel-air mixture entering the catalyst. This pre-
heat combustor is not required for inlet air temperatures above
650 K. Main air is introduced immediately downstream of the
preheat burner. The multipoint fuel injector is an arrangement
tor is operating perilously close to the lean blowout limit. One of six venturis surrounding a seventh at the center. Gas tempera-
method of extending the lean blowout limit down to lower tures entering the catalyst bed are in the range from 644 to 925
equivalence ratios is by incorporating a combustion-enhancing K. The catalytic reactor is a multiple-piece bed design using a
catalyst within the combustor. Catalytic combustion is a ceramic honeycomb substrate coated with catalytically active
"flameless" process, which allows fuel oxidation to take place materials. Its function is to initiate reactions of the main fuel-
at temperatures well below the normal lean flammability limits air mixture. Downstream of the catalyst bed is a postcatalyst
of the fuel-air mixture. For this reason, the use of catalysts in combustion zone, which is stabilized behind a bluff-body
gas turbine combustion to replace part of the thermal reaction flameholder. This postcatalyst combustor demonstrated efficient
zone allows stable combustion to occur at peak temperatures CO and UHC burnout. Short duration rig testing at 9 atm pres-
that are around 1000 K lower than those of conventional com- sure indicated NOn emissions below 5 ppm. The system now
bustors. Combustion at such reduced temperatures can be ex- being developed at Solar for engine application includes fuel
pected to decrease the production of thermal NO~ dramatically. staging and variable geometry.
A schematic diagram to illustrate the principle of catalytic The advantage of using a postcatalyst combustor to complete
combustion is shown in Fig. 51, Fuel is injected upstream of the combustion process is that it allows the catalytic bed to
the reactor to vaporize and mix with the inlet air. The fuel-air operate at temperatures lower than the combustor outlet temper-
mixture then flows into a catalyst bed, or reactor. This reactor ature. This is a significant advantage because it improves the
may consist of several sections, each made of a different kind reliability and extends the life of the catalyst bed. Ozawa et al.
of catalyst. It is desirable to use a catalyst that is active at low (1994a, b) used this approach in designing a combined cata-
temperatures at the inlet, while subsequent sections need to lytic/premix combustor to operate at a combustor outlet temper-
be selected for good oxidation efficiency. Downstream of the ature of 1570 K while keeping the catalyst bed temperature
catalytic bed, a thermal reaction zone may be provided to permit down to 1270 K. The system is shown schematically in Fig.
catalytically initiated reactions to continue. Such reactions can 53. It consists of an annular prebumer, six catalytic combustor
contribute as much as 20 percent of the total temperature rise segments, six premixing nozzles, a premixed combustion sec-
achieved in the combustor. tion downstream of the catalyst, and an air bypass valve. Inlet
One problem with catalytic combustors is a tendency toward air is heated to 720 K by the preburner and is distributed to
autoignition of the fuel upstream of the catalyst. Although the both catalytic segments and premixing nozzles. The premixed,
equivalence ratios of interest are well below the lean flammabil- gaseous fuel-air mixture that flows around the catalyst seg-
ity limit and, in theory, should not be susceptible to autoignition, ments is ignited by the hot combustion products emanating from
locally richer mixtures exist near the fuel injector before com- the catalysts, and lean premix combustion then occurs to raise
plete mixing has taken place. Thus, it is essential that mixing the combustor outlet temperature up to 1570 K. The function
be achieved in less than the autoignition delay time. Optimum of the bypass valve is to add air to the combustor efflux gases
catalyst performance also requires the inlet fuel-air mixture to at part-load operation.
be completely uniform in regard to temperature, composition, Measurements of NOx at atmospheric pressure when burning
and velocity profile, since this assures effective use of the entire natural gas were obtained by varying the fuel flow rate and
catalyst area and prevents damage to the substrate due to local bypass valve opening to maintain the catalyst bed at a constant
high gas temperatures. Thus the primary function of the fuel temperature. NO~ emissions were always below 10 ppm, of
preparation process is to provide complete and uniform mixing
of fuel and air before entry into the catalyst. Rudimentary tests
carried out by Anderson et al. (1982) showed that approxi-
mately 70 percent fuel vaporization is required for stable cata-
lyst operation when burning liquid fuels. These workers used
multiple air-assist atomizers to achieve good atomization and
rapid fuel evaporation at turndown ratios in excess of ten to
one. This multipoint fuel injection enabled complete fuel vapor-
ization and mixing to occur in a mixing zone with a length/
diameter ratio of five.
Cowell and Larkin (1994) have conducted subscale testing
on several different designs of catalytic combustors. The most PREMIXFED
advanced concept is shown schematically in Fig. 52. It is de- [ unit =mm ]
signed to operate at lean conditions to suppress NOn emissions.
CO and UHC emissions are of less concern because they are Fig. 53 Catalytic combustor for natural gas

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 647

Downloaded From: on 01/27/2016 Terms of Use:

which a significant proportion came from the preburner. The
total pressure loss of the combustor was around 3 percent, which
is lower than the values normally associated with conventional
The notion of completing the combustion process in the post-
catalyst region to avoid exposure of the catalyst to the high
temperatures responsible for deactivation and thermal shock
fracture of the supporting substrate has been pursued by other I
workers, including Dalla Betta et al. (1994). In their combustor, Fig. 54 Combination of catalytic and conventional staged combustor
natural gas mixed with air is supplied to the catalyst. Combus-
tion is initiated by the catalyst and is completed by homoge-
neous reactions downstream. The potential of this approach has
been demonstrated in subscale rig tests at pressures up to 14 combustion over the broad range of conditions encountered in
atm and temperatures above 1500 K. Measured emissions were modern gas turbines.
NOx < 1 ppm, CO < 2 ppm and UHC < 2 ppm, thus demon- With variable geometry, as the fuel flow increases with load,
strating the capability of achieving ultralow NOx along with proportionally larger amounts of air are diverted through the
low CO and UHC. catalyst to keep its reaction temperature within the allowable
The catalyst employed by these workers required an inlet air limits. The remaining air bypasses the catalyst and dilutes the
temperature of 720 K, which necessitated the use of a preburner. combustion gases downstream. Fuel staging can be accom-
However, it is reported that recent catalyst designs have reduced plished, for example, by combining a conventional combustor
the required inlet temperature to the point where the catalyst with a catalytic combustor, as illustrated in Fig. 54. The com-
can be operated at the compressor discharge temperature for bustion requirements at low load conditions are met using a
some machines. conventional burner. Then, at full-load conditions, where the
• The very low emissions levels achieved by Dalla Betta et al. overall equivalence ratio is compatible with catalytic combus-
may be due in part to the high degree of uniformity produced by tion, most or all of the required temperature rise is achieved
the fuel-air mixing system used in the test rig, which featured a via catalytic combustion. This method of staging requires a
static mixer located between the fuel injector and the catalyst. careful control of air splits to provide the required catalyst
Vortmeyer et al. (1994) also integrated static mixers into the pressure drop. This may necessitate the use of variable geometry
premixing/prevaporizing section upstream of the catalyst to in addition to fuel staging to control the air flows to the conven-
minimize imperfections in fuel-air mixing. Their combustor tional and catalyst stages.
design also featured a thermal reactor downstream of the cata- Most of the work now in progress on catalytic combustion
lyst to allow it to operate at temperatures well below the maxi- for gas turbines is directed toward stationary applications. For
mum flame temperature without the need for tertiary fuel injec- aircraft engines, the concerns expressed above regarding dam-
tion into the thermal reactor. This was achieved by employing age to the catalyst when operating for long periods at high
very short catalyst segments and/or high flow velocities. The pressures and temperatures would appear to put it out of con-
test combustor was supplied with three different groups of tention at this time, especially as both LPP and RBQ concepts
fuels--paraffins, alcohols, and commercial liquid fuels. The show almost equal promise with less risk.
range of operation was limited by the requirements of at least
99.9 percent combustion efficiency at the end of the thermal Correlation and Modeling of NOx and CO Emissions
reactor to minimize the emissions of CO and UHC. Catalyst
temperatures never exceeded 1300 K up to a maximum gas Empirical Models. Ever since the gas turbine became rec-
temperature at the combustor exit of 1670 K. The overall com- ognized in the late 1960s as a significant contributor to atmo-
bustor pressure loss was 6 percent, which is just within the spheric pollution, efforts have been made to develop simple
accepted range for conventional combustors. The catalyst of 40 models for correlating experimental data on exhaust emissions
mm length was responsible for almost half of this loss (2.35 in terms of all the relevant parameters. These include combustor
percent). The test program was conducted at atmospheric pres- operating conditions, combustor design features, fuel type, and
sure and was characterized by remarkably low NOx emissions fuel spray characteristics.
levels. For propane and methanol no measurable NOx was Empirical models can play an important role in the design
formed. The combustion of Diesel oil yielded NOx emissions and development of low-emissions combustors. They serve to
below 2 ppm. reduce the complex problems associated with emissions to
A main drawback to the application of catalytic combustors forms that are more meaningful and tractable to the combustion
in modern gas turbines is the limited life expectancy of catalysts engineer. They also permit more accurate correlations of emis-
under typical high-temperature conditions. Most combustion rig sions for any one specific combustor than can be achieved by
testing to date has involved relatively short test periods of less the far more comprehensive models discussed below.
than 100 hours. However, lifetimes of around 5000 hours are In attempting to derive an empirical model for emissions,
required for engine applications. It is feared that serious catalyst emphasis is placed on NOx and CO. For both these species it
degradation could occur when operating at high pressures and may be assumed that their exhaust concentrations are dependent
temperatures for long durations. Another major drawback is on three terms, which are selected to represent the following:
that although they are capable of achieving levels of combustion ( 1 ) mean residence time in the combustion zone, (2) chemical
efficiency very close to 100 percent, this occurs only over a reaction rates, and (3) mixing rates. Expressions for these three
fairly narrow range of operating temperatures. The lower tem- parameters can be derived in terms of combustor size, liner
perature bound is set by lean blowout limits; the upper bound pressure loss, airflow proportions, and operating conditions of
is determined by catalyst and substrate material limitations. inlet pressure, temperature, and air mass flow rate. We have
Depending on the type of catalyst employed, stable combustion Residence time = L / U ~ P V c / m A T (17)
can be sustained over a temperature range of only a few hundred
degrees Kelvin without serious loss of performance. This corre- Reaction rate ~ P" exp a T (18)
sponds to a range of overall fuel/air ratios of around 1.4 to 1,
Mixing rate ~ ( A P / P ) x (19)
as opposed to the 5 to 1 readily achieved with conventional
combustors. It is essential therefore that catalytic combustion be where L = liner length, U = liner flow velocity (average), V,
combined with variable-geometry and/or fuel staging to sustain = combustion volume, P = pressure, T = gas temperature, ma

648 / Vol. 117, OCTOBER 1995 Transactions of the A S M E

Downloaded From: on 01/27/2016 Terms of Use:

= air flow rate, ~xP = liner pressure drop, and a, n, and x are
constants to be determined experimentally. By combining these
terms we obtain for NO,,
0 /
3C 2 zx
NO~ =f(residence time)(reaction rate)(mixing rate) (20) 3
From analysis of experimental data for several different com- 5 v
bustors (Lefebvre, 1984), values of a, n, and x were established 6 x xx
and substituted into Eq. (20) to obtain
NOx = 9 × 10-8plZsV~ exp(O.O1T, t)/maTp~ g/kg fuel (21) "~ 2o
where P is the combustion pressure in kPa and Vcis the combus-
tion volumne in m 3. The effect of variations in overall combus-

tot fuel/air ratio is included via its influence on primary-zone

temperature. Fuel type affects both flame temperature and mean z
drop size. For NOx, drop size is less important since, at the
high pressure conditions where NOx emissions are most promi- ,0 FIOI
nent, the fraction of the total combustion volume employed in / P3 = 3 8 6 - 1 2 6 6 kPo
fuel evaporation is so small that variations in fuel drop size
have only a small effect on NO~ emissions. For this reason, / T 3 = 4 6 4 - 850 K
most correlating parameters for NOr ignore the effects of mean H~A= 8.5- 20.5 kg/s
drop size.
Equation (21) takes account of the fact that, in the combus- 0 i I~3 i i I I
tion of heterogeneous fuel-air mixtures, it is the stoichiometric 20 30
flame temperature, T~,, that determines the formation of NOt. NOn (predicted), g/kg
However, for the residence time in the combustion zone, which
is also significant to NOn formation, the appropriate temperature Fig. 56 Comparison of measured and predicted values of NO,, for a GE
term is the average value Tp~, as indicated in the denominator F101 combustor
of Eq. (21).
Equation (21) is suitable for conventional spray combustors where P is the combustion pressure in Pa, T is the combustion
only. For lean premix/prevaporize combustors in which the temperature in degrees Kelvin, and t is the NOx formation time
maximum attainable temperature is Try, it may still be used, in ms. For aircraft combustors, Odgers and Kretschmer quote
provided that Tpz is substituted for T,. The excellent correlation formation times of 0.8 ms (airblast atomizers) and 1.0 ms (pres-
of experimental data on NOw provided by Eq. (21) for GE J79- sure atomizers). For industrial combustors burning liquid fuels,
17A and F101 combustors is illustrated in Figs. 55 and 56, quoted formation times range from 1.5 to 2.0 ms.
respectively. Many other semi-empirical models for predicting
Lewis (1991).
NO~ emissions have been derived. For a critical review of mod-
els developed before 1980, reference should be made to Mellor NOx = 3.3192 × 10 -6 exp(0.0079776T)P °5 ppmv (23)
( 1981 ). Some of the more recent expressions for NOx emissions where P is the combustion pressure in atm. This equation is
include the following: intended to show the amount of NOx formed in lean, homoge-
Odgers and Kretschmer (1985) neous combustion. It suggests that NO~ formation depends only
on the postcombustion temperature and pressure and is com-
NOr = 29 exp - (21,670/T)P °'66 pletely independent of the residence time of the gases in the
combustor. According to Lewis (1991), "the time involved is
X [1 - exp - (250t)] g/kg fuel (22)
not the normal residence time of the combustion products, but
rather the relaxation time of the molecules involved, primarily
the nitrogen molecule, and thus, is the same in all combustion
.50 systems using air." This finding appears to be at variance with
Fuel Symbol
published data on the effect of residence time on NOx formation
I o
2 zx (see, for example, Rizk and Mongia, 1990; Nicol et al., 1992;
5 0 Zelina and Ballal, 1994), all of which indicate that NOx concen-
4 0 trations increase with residence time. However, these workers
5 v also noted that the influence of residence time on NOx formation
6 x
2C x V diminishes with reduction in equivalence ratio, the effect being
° relatively small for very lean mixtures, as studied by Lewis.
-3 Some expressions for NOx include time directly, as in Eq. (22),
but in others it is embodied into the expression via a mass flow
E rate term, as in Eq. (21). However, even expressions for NOx
that take no account of residence time whatsoever (Eq. (23),
o~ IC for example) can nevertheless provide an excellent prediction
of experimental data because the residence time of all aero
P3 = 251-1374 kPa gas turbine combustors tends to be the same at around a few
T5 = 413- 787 K milliseconds. It is only when expressions derived for industrial
gas turbine combustors having much longer residence times are
rhA= 1.7-7.4 kg/e
applied to aero combustors, or vice versa, that the lack of a
I I term for residence time becomes important.
I0 20 ~0
NOx (predicted), g/kg Rokke et al. (1993)
NOx = 18.1pl42m°x3q °'72 ppm (24)
Fig. 55 Comparison of measured and predicted values of NOwfor a GE
J79-17A combustor where P is the combustion pressure in atm, ma is the combustor

J o u r n a l of E n g i n e e r i n g f o r G a s T u r b i n e s a n d P o w e r OCTOBER 1995, Vol. 117 / 6 4 9

Downloaded From: on 01/27/2016 Terms of Use:

air flow in kg/s, and q is the fuel/air ratio. This equation was
found to correlate very satisfactorily measurements of NOx FIO0
emissions from five different natural gas-fired machines op-
All Fuels
erating in the power range from 1.5 to 34 MW. Although com-
bustion temperature is conspicuous by its absence in Eq. (24),
its influence on NO~ emissions is accounted for by the inclusion
of a fuel/air ratio term.

Rizk and Mongia (1994)

NOn = 15.10J4(t - 0.5t,) °'5 exp
- (71,100/7~,)P-°'°5(AP/P) -°'5 g/kg fuel (25)

where t is the primary-zone residence time in seconds, te is the 8

fuel evaporation time, and ~xP/P is the nondimensional liner
pressure drop. An interesting feature of this equation is that it
includes a term, t,, to account for the effects of fuel drop size P5=370 - 1520 kPa
on NO, emissions. According to Eq. (25), a reduction in mean 0 q-4 -864 K
drop size should increase NOx emissions by reducing the time
required for fuel evaporation. However, if combustion takes aliA=2.8- 73 kg/s
place under conditions where evaporation times are negligibly
small in comparison with the total combustor residence time,
1.0 2 5 I0 20 50
NO, emissions can actually go down with reduction in mean
CO (predicted), g/kg
drop size, as observed by Rink and Lefebvre (1989).
For more information on semi-analytical equations for the Fig. 57 Comparison of measured and predicted values of CO
estimation of NOn emissions from gas turbines, reference should
be made to Becker and Perkavec (1994) and Nicol et al. (1994).
Comparisons between these different equations for predicting
NO, emissions from conventional combustors is prohibited by CO = 0.18 × 109 exp(7800/Tpz)/
the fact that in some cases the units for NOn are in parts per PZ(t - 0 . 4 t ~ ) ( A P / P ) °'5 g/kg fuel (28)
million and in others grams per kilogram of fuel. These units
cannot be interchanged unless the equivalence ratio is also where t~ is the fuel evaporation time. This equation exhibits a
known. However, all of these expressions for NOx provide an slightly lower dependence on combustion temperature and a
excellent fit to the experimental data employed in their deriva- slightly higher dependence on pressure than Eq. (26).
tion. This is perhaps hardly surprising since the formation of Correlating parameters have also been developed for UHC
NOx is largely dominated by one parameter o n l y - - t h e maxi- and smoke. The interested reader is referred to Rizk and Mongia
mum temperature in the combustion zone. (1994) for the latest work.
Similar correlations to those presented above for NO~ have
Numerical Models, The application of numerical models
been developed for CO. One important difference stems from
to the design of gas turbine combustors owes its origin to the
the fact that the formation of CO in the primary combustion
pioneering work of several workers, prominent among whom
zone takes appreciably longer than the time required to produce
are Jones and Priddin (1978), Mongia (1987), and Serag-Eldin
NO~. In consequence, the relevant temperature is not the local
and Spalding (1979).
peak value adjacent to the evaporating fuel drops, but the aver-
age value throughout the primary zone, Tpz. Also, because CO In recent years, the detailed representation of combustor flow
emissions are most important at low pressure conditions, where fields made available through numerical methods has been used
to provide some insight into the interaction of swirling and
evaporation rates are relatively slow, it is necessary to reduce
the combustion volume V~by the volume occupied in fuel evap- recirculating flows with fuel sprays and the effect of this interac-
oration V~. tion on atomization, dispersion of fuel drops, evaporation, mix-
ing, and turbulent combustion. Mongia and co-workers, in par-
According to Lefebvre (1984) we have
ticular, have made extensive use of numerical methods to obtain
CO = 86 m Tp~ exp - (0.00345Tpz)/ a better understanding of the formation of various pollutant
species within the combustor, while the use of computational
(V~ - V,)( AP/P)°'SP 1'5 g/kg fuel (26) combustor dynamics (CCD) to provide guidance to the combus-
tion engineer during the design and development process has
where V~, the volume employed in fuel evaporation, is given become an accepted practice in many gas turbine companies
by (e.g., Correa and Shyy, 1987; McGuirk and Palma, 1993; Prid-
din and Coupland, 1986; and Sturgess, 1986).
Ve = 0.55mpzD o~/ ppzh~ff (27) At the present time much of the computer modeling in gas
turbine combustors is accomplished using versions of the
where m~,z is the primary-zone airflow rate, Do is the Sauter TEACH code and the k - e turbulence model. It seems likely
mean diameter of the fuel spray, pp~ is the density of the primary- that TEACH-based codes and their derivatives, along with the
zone gas, and hee~is the effective evaporation constant (Chin commercially available FLUENT code, will continue to play
and Lefebvre, 1982). In this equation it is of interest to note useful roles for some time to come, with updates to the codes
that Ve is proportional to the square of the initial mean drop being made in the light of new experimental data. However, it
size. This highlights the importance of good atomization to the must be stated that the three-dimensional comprehensive models
attainment of low CO. The ability of Eq. (26) to predict CO now in vogue are costly in computer time and the results ob-
emissions from a P&W F100 and a GE F101 combustor is tained often fall short of expectations. As Rizk and Mongia
illustrated in Figs. 57 and 58, respectively. (1993a) have pointed out, due to the incomplete understanding
A similar form of expression to Eq. (26) has been derived of the various combustion processes, analytical calculations on
by Rizk and Mongia (1994) as practical gas turbine combustors cannot yet be considered quan-

650 / Vol. 117, OCTOBER 1995 Transactions of the A S M E

Downloaded From: on 01/27/2016 Terms of Use:

tor has just received FAA certification on Airbus Industrie A320
0 and A321 aircraft.
F I01 At the present time, lean premixed combustion appears to be
All Fuels
the only technology available for achieving sub-10 ppm NOx
emissions in practical combustors. However, such low NOx lev-
els can only be attained if the fuel and air supplied to the
combustion zone are almost perfectly premixed. At high power
O0 conditions, where combustor inlet air temperatures can ap-
20 proach 1100 K, most hydrocarbon fuels readily ignite on contact
with air. Thus the problems of spontaneous ignition and flash-
back need to be fully addressed before lean premixed combus-
tion can be applied with confidence to aircraft engines. Another

!'°f problem associated with well-mixed combustion systems is that

of acoustic resonance, which occurs when the combustion pro-
cess becomes coupled with the acoustics of the combustor. This
problem could be of great importance to the future development
of lean prernixed combustors.
Other main contenders as ultralow-NOx combustors are the
P3= 586-1266 kPo rich-burn, quick-quench, lean bum (RQL) and catalytic com-
T3 = 4 6 4 - 8 5 0 K bustion concepts. The work carried out so far on various RQL
configurations has shown that this concept has considerable
= 8.5-20,5 kg/s
promise for very low NOx emissions. Its future prospects depend
I largely on whether the rich combustion products emanating
I 2 5 I0 20 50 I00 from the primary zone can be mixed with the remaining com-
co (predicted), g/kg
bustion air quickly enough to avoid the creation of high-temper-
Fig. 58 Comparison of meas'ured and predicted values of CO for a GE ature, NOx-forming regions. Catalytic combustors allow very
F101 combustor low combustion temperatures, which ensure exceptionally low
levels of NOx. Their main drawback is the limited life expec-
tancy of catalysts and substrates under typical high temperature
conditions. The use of postcatalyst combustion allows the cata-
titatively accurate. Significant advances in physical submodels,
lytic bed to operate at lower temperatures with consequent bene-
numerics, and in the simulation of actual hardware are needed
fits in terms of reliability and longer life. Most of the work
to improve the predictive capability of current models (Peters,
now in progress is directed toward catalytic combustors for
1988; Mongia, 1993).
stationary engines. Considerable advances in high-temperature
Prompted by the limitations of current CCD models, and
materials will be needed to raise their reliability to the standard
encouraged by the long history of success in routine applications
required for aircraft engines.
of simple correlations to combustor design and development,
The continuing pressure to conserve fuel resources can only
Rizk and Mongia (1986) proposed a hybrid modeling approach
be met by raising the engine cycle efficiency. This approach
that combines comprehensive modeling with physically realistic
yields an important additional bonus in the form of less COz
and well-established expressions for estimating various im-
but it also results in higher combustion temperatures and higher
portant performance parameters such as combustion efficiency,
levels of NOx. Thus, the need to burn less fuel and thereby
lean blowout limits, pattern factor, liner wall temperatures and
generate less CO2 is in direct conflict with the equally important
gaseous pollutant emissions. This empirical/analytical approach
need to reduce NOx. For the forseeable future it seems likely
to combustor design has been successfully applied to 22 gas
that engine pressure ratios will rise up to a maximum value of
turbine combustors (Mongia, 1993). It represents a logical step
around 50, beyond which any further increase will depend on
forward that could prove of great value to combustion engineers
the outcome of current research and development efforts in
in providing quantitative design guidance until such time as
ultralow NOx combustion.
new knowledge on turbulent combustion models, numerics, grid
We are now entering into an area of increasing competitive-
resolution, and boundary conditions become available for im-
ness between different engine manufacturers in regard to low-
proving the accuracy of current CCD models. Examples of the
NOx combustors. For stationary gas turbines, the attainment of
use and application of hybrid models can be found in Rizk and
low-NOx has become the foremost marketing issue, thus offer-
Mongia (1986, 1993b, c, 1994).
ing important opportunities to those engine manufacturers who
are willing to commit large resources to pollutants reduction
Concluding Remarks Fuel reserves are notoriously difficult to predict but, for the
It is fortunate that the gas turbine not only offers the highest next twenty years or thereabouts, kerosine-type fuels should be
thermal efficiency as a prime mover for aircraft propulsion and fairly readily available for aircraft engines, while natural gas
power generation for the foreseeable future, but also has the will continue to be the favored fuel for baseload operations of
potential for the lowest possible levels of pollutant emissions. stationary engines. With the ultimate depletion of conventional
Most of the drive toward more strict control of pollutant emis- hydrocarbon fuels and natural gas, oils derived from coal,
sions from gas turbines is being directed at oxides of nitrogen shales, and tar sands will assume increasing importances. Such
(NOx). Low NOx levels are readily achieved by lowering the fuels are known to be of higher viscosity and lower volatility
peak flame temperature within the combustor. The challenge is than current light-distillate fuels, thus exacerbating the problems
to keep flame temperatures down at high-power conditions with- of atomization and fuel prevaporization.
out incurring unacceptable penalties in combustion performance Fuel preparation is of vital importance in the attainment of
when operating at low power conditions. For the immediate low pollutant emissions. Future advances in low-emissions
future, the introduction into service of staged combustion ap- combustion technology will demand a detailed knowledge of
pears to be most promising, despite its attendant penalties for the properties and structure of the sprays produced in airblast
the engine in terms of more complex fuel scheduling and control and pressure atomization. The properties of prime interest in-
strategy. In fact, a General Electric dual-annular staged combus- clude drop size distributions, droplet and gas velocities, droplet

J o u r n a l of E n g i n e e r i n g f o r G a s T u r b i n e s a n d P o w e r OCTOBER 1 9 9 5 , Vol. 1 "17 / 651

Downloaded From: on 01/27/2016 Terms of Use:

trajectories, and droplet mass flux. Such detailed information is Appleton, J. P., and Heywood, J. B., 1973, "The Effects of Incomplete Fuel-
Air Mixing in a Burner on NO Formation From Nitrogen in the Air and in the
essential for the successful modeling of sprays, but the extent Fuel," Fourteenth Symposium (International) on Combustion, The Combustion
to which predictions based on spray models can be utilized in Institute, Pittsburgh, PA, pp. 777-786.
combustor design depends largely upon the degree to which Bahlmann, F. C., and Visser, B. M., 1994, "Development of a Lean-Premixed
they can be validated or refuted by accurate experimental data. Two-Stage Annular Combustor for Gas Turbine Engines," ASME Paper No. 94-
This situation calls for more collaborative efforts between the
Bahr, D. W., 1982, "HC and CO Emission Abatement via Selective Fuel
modeling and experimentation communities. Injection," ASME Paper No. 82-GT-178.
Although considerable insight has been gained into the inter- Bahr, D. W., 1987, "Technology for the Design of High Temperature Rise
nal flow patterns and external spray characteristics of both pres- Combustors," AIAA Journal of Propulsion and Power, Vol. 3, No. 2, pp.
sure-swirl and airblast atomizers, considerable work remains to 179-186.
Bahr, D. W., 1991, "Aircraft Engines NOx Emissions--Abatement Progress
be done. Except for atomization occurring at impractically low and Prospects," Proceedings of Tenth International Symposium on Air-Breathing
velocities, there is still no basic method for predicting the drop Engines, Nottingham, England.
size distributions that will result from the disintegration under Bahr, D. W., 1992, "Aircraft Turbine Engine NOx Emission Limits--Status
controlled conditions of a liquid jet or sheet. It has been known and Trends," ASME Paper No. 92-GT-415.
Bahr, D. W., 1995, Lecture notes for short course on gas turbine combustion;
for more than thirty years that the geometry, size, shape, and held at the University of California, Irvine, Mar. 13-17.
surface finish of the flow passages within an atomizer have a Bauckhage, K., 1988, "The Phase-Doppler Difference Method. A New Laser-
strong influence on the internal flow patterns and the external Doppler Technique for Simultaneous Size and Velocity Measurement," Particles
spray characteristics, but these important effects are not yet and Particle System Characterization, Vol. 5, pp. 16-22.
Bayle-Labour6, G., 1991, "Pollutant Emissions From Aircraft Engines: a Situa-
sufficiently well understood to form an integral part of the de- tion Under Control," Revue Scientifique SNECMA, 2rid ed.,
sign process. The widespread use of laser diagnostic techniques Becker, T., and Perkavec, M. A., 1994, "The Capability of Different Semiana-
in spray characterization has created a rapidly growing data lytical Equations for Estimation of NOx Emissions of Gas Turbines," ASME
base that provides theoreticians and modelers with new opportu- Paper No. 94-GT-282.
Chert, S. K., Lefebvre, A. H., and Rollbuhler, J., 1992, "Factors Influencing
nities for establishing a scientific basis for a methodical ap- the Spray Cone Angle of Pressure-Swirl Atomizers," ASME JOURNALOF ENGI-
proach to the design of fuel injectors and fuel-air mixing sys- NEERINGFOR GAS TURBINESAND POWER, Vol. 114, pp. 97-103.
tems. Chen, S. K., Lefebvre, A. H., and Rollbuhler, J., 1993, "Factors Influencing
A drawback to advanced laser diagnostics and comprehensive the Circumferential Liquid Distribution From Pressure-Swirl Atomizers," ASME
modeling is that because they are both challenging and intellec- JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER, Vol. 115, pp.
tually rewarding in their own right it is easy to forget that they Chigier, N., 1983, "Drop Size and Velocity Instrumentation," Prog. Energy
are only means to an end. This author is not aware of any Combust. Sci., Vol. 9, pp. 155-177.
significant advance in atomizer or combustor technology that Chigier, N., and Stewart, G., 1984, "Particle Sizing and Spray Analysis,"
can be attributed to laser diagnostics or comprehensive model- Optical Engineering, Vol. 23, No. 5, pp. 554-640.
Chin, J. S., and Lefebvre, A. H., 1982, "Effective Values of Evaporation
ing. All too often these activities are pursued to the detriment Constant for Hydrocarbon Fuel Drops," Proceedings of the 20th Automotive
of innovation and creativity, which are the two essential ingredi- Technology Development Contractor Coordination Meeting, pp. 325-331.
ents for progress in these areas. Chin, J. S., Lefebvre, A. H., and Sun, F. T.-Y., 1992, "Temperature Effects
AND POWER, Vol. 114, pp. 353-358.
Acknowledgments Chin, J. S., and Lefebvre, A. H., 1993, "Influence of Flow Conditions on
Deposits From Heated Hydrocarbon Fuels," ASME JOURNALOF ENGINEERING
Any paper that attempts to review a given topic must neces- FOR GAS TURBINESAND POWER, Vol. 115, pp. 433-438.
sarily draw heavily on the work of others and this paper is no Chin, J. S., 1994, ',An Engineering Calculation Method for Multi-component
Stagnant Droplet Evaporation With Finite Difthsivity," ASME Paper No. 94-
exception in this regard. If the author's name appears more GT-440.
often than is merited by his contribution to the subject, this is Chin, J. S., 1995a, "Advanced Droplet Evaporation Model for Turbine Fuels,"
due in large measure to his desire to recognize the valuable AIAA Paper No. 95-0493.
contributions made by his various co-authors of many publica- Chin, J. S., 1995b, "An Engineering Calculation Method for Turbine Fuel
tions in the areas of fuel injection and gas turbine combustion, Droplet Evaporation at Critical Conditions With Finite Liquid Diffusivity," AIAA
Paper No. 95-0494.
most of whom were his students either at Cranfield University Claeys, J. P., Elward, K. M., Mick, W. J., and Symonds, R. A., 1993, "Combus-
in the U.K. or at Purdue University in the U.S.A. tion System Performance and Field Test Results of the MS7001F Gas Turbine,"
I would like to express my gratitude to the IGTI Gas Turbine ASME JOURNALOF ENGINEERINGFOR GAS TURBINESAND POWER, Vol. 115, pp.
Scholar Program and to Professor John Denton, the Chairman 537-546.
Correa, S. M., and Shyy, W., 1987, "Computational Models and Methods for
of the IGTI Scholar Award Review Committee, for his encour- Continuous Gaseous Turbulent Combustion," Prog. Energy Combust. Sci., Vol.
agement and support during the preparation of this paper. I 13, pp. 249-292.
would also like to thank my friends and colleagues in the gas Correa, S. M., 1991, "Lean Premixed Combustion for Gas Turbines: Review
turbine combustion community for their confidence in nomi- and Required Research," Fossil Fuel Combustion, ASME PD-Vol. 33.
nating me for this award. Cowell, L. H., and Larkin, M. P., 1994, "Development of a Catalytic Combus-
tor for Industrial Gas Turbines," ASME Paper No. 94-GT-254.
Custer, J. R., and Rizk, N. K., 1988, "Influence of Design Concept and Liquid
Properties on Fuel Injector Performance," AIAA Journal of Propulsion and
References Power, Vol. 4, No. 4, pp. 378-384.
Aigner, M., and Muller, G, 1993, "Second Generation Low-Emission Combus- Dalla Betta, R. A., Schlatter, J. C., Nickolas; S. G., Yee, D. K., and Shoji,
tots for ABB Gas Turbines: Field Measurements With GTllN-EV," ASME T., 1994, "New Catalytic Combustor Technology for Very Low Emissions Gas
JOURNAL OF ENGINEERINGFOR GAS TURBINESAND POWER, Vol. 115, pp. 533- Turbines," ASME Paper No. 94-GT-260.
536. Davis, L. B., and Washam, R. M., 1989, "Development of a Dry Low NOw
Altemark, D., and Knauber, R., 1987, VDI Berichte No. 645, pp. 299-311. Combustor," ASME Paper No. 89-GT-255.
Anderson, D. N., 1975, "Effects of Equivalence Ratio and Dwell Time on De Corso, S. M., and Kemeny, G. A., 1957, "Effect of Ambient and Fuel
Exhaust Emissions From an Experimental Premixing Prevapodzing Burner," Pressure on Nozzle Spray Angle," Trans. ASME, Vol. 79, pp. 607-615.
ASME Paper No. 75-GT-69. Delarosa, A. B., Sankar, S. V., Wang, G., and Bachalo, W. D., 1993, "Particle
Anderson, S. J., Friedman, M. A., Kdll, W. V., and Kesseldng, J. P., 1982, Diagnostics and Turbulence Measurements in a Confined Isothermal Liquid
"Development of a Small-Scale Catalytic Gas Turbine Combustor," ASME JOUR- Spray," ASME JOURNALOF ENGINEERINGFOR GAS TURBINESAND POWER, Wol.
NAL OF ENGINEERINGFOR GAS TURBINESAND POWER, Vol. 104, pp. 52-57. 115, pp. 499-506.
Angello, L., and Lowe, P., 1989, "Dry Low NOx Combustion Development Desaulty, M., 1991, "Turbine Engine Combustor Design at SNECMA," Pro-
for Electric Utility Gas Turbine Applications; A Status Report," ASME Paper ceedings of International Symposium on Air-Breathing Engines, ISABE 91-7064,
No. 89-GT-254. pp. 80-90.
Aoyama, K., and Mandai, S., 1984, "Development of a Dry Low NOx Combus- Dittman, R., Feld, H. J., Samenfink, W., and Wittig, S., 1992, "Multiple Wave-
tor for a 120 MW Gas Turbine," ASME JOURNAL OF ENGINEERINGFOR GAS length Extinction Technique for Particle Characterization in Dense Particle
TURBINES AND POWER, Vol. 106, pp. 795-800. Clouds," Nuremburg, Germany, Preprint 2, pp. 507-518.

652 / Vol. 117, OCTOBER 1995 Transactions of the ASME

Downloaded From: on 01/27/2016 Terms of Use:

Ducourneau, F., 1974, "Inflammation Spontanee de Melanges Riches Air- for Advanced Aircraft Engines, AGARD Conference Proceedings 536, pp.
Kerosine," Entropie, No. 59. 21-t-10.
Durst, F., Melling, A., and Whitelaw, J., 1981, Principles and Practice of McDonell, V. G., Lee, S. W., and Samuelson, G. S., 1994a, "Spray Behavior
Laser-Doppler Anemometry. in Reacting and Non-reacting Flow Fields Downstream of an Aero-engine Com-
Etheridge, C. J., 1994, "Mars Solo NOx Lean Premix Combustion Technology bustor Dome," Twenty-Fifth Symposium (International) on Combustion, The
in Production," ASME Paper No. 94-GT-255. Combustion Institute, Pittsburgh, PA.
Fennimore, C. P., 1971, "Formation of Nitric Oxide in Premixed Hydrocarbon McDonell, V. G., Seay, J. E., and Samuelson, G. S., 1994b, "Characterization
Flames," Thirteenth Symposium (International) on Combustion, The Combustion of the Non-reacting Two-Phase Flow Downstream of an Aero-engine Combustor
Institute, Pittsburgh, PA, pp. 373-380. Dome Operating at Realistic Conditions," ASME Paper No. 94-GT-263.
Flanagan, P., Gretsingir, K., Abbasi, H. A., and Cygan, D., 1992, "Factors McGuirk, J. J., and Palma, J. M. L. M., 1993, "The Flow Inside a Model Gas
Influencing Low Emissions Combustion," ASME PD-Vol. 89, Fossil Fuels Com- Turbine Combustor: Calculations," ASME JOURNALOF ENGINEERINGFOR GAS
bustion. TURBINES AND POWER, Vol. 115, pp. 594-602.
Fric, T. F., 1992, "Effects of Fuel-Air Unmixedness on NO, Emissions," McKnight, D., 1978, "Development of a Compact Gas Turbine Combustor to
AIAA Paper No. 92-3345. Give Extended Life and Acceptable Exhaust Emissions," ASME JOURNALOF
Goodger, E. M., and Eissa, A. F. M., 1987, "Spontaneous Ignition Research; ENGINEERINGFOR POWER, Vol. 100, July.
Review of Experimental Data," Journal of the Institute of Energy, Vol. 84, pp. McVey, J. B., Russell, S., and Kennedy, J. B., 1987, "High-Resolution Pattema-
84-89. tor for the Characterization of Fuel Sprays," AIAA Journal of Propulsion and
Heberling, P. V., 1976, "Prompt NO Measurements at High Pressures," Six- Power, Vol. 3, No. 3, pp. 202-209.
teenth Symposium (International) on Combustion, The Combustion Institute, McVey, J. B., Kennedy, J. B., and Russell, S., 1989, "Application of Advanced
Pittsburgh, PA, pp. 159-168. Diagnostics to Airblast Injector Flows," ASME JOURNALOF ENGINEERINGFOR
Hebrard, P., Trichet, P., and Bardey, X., 1991, "Experimental Investigation of GAS TURBINESAND POWER, VOI. 111, No. 1, pp. 53-62.
Two-Phase Flow in the Near Field of an Airblast Atomizer," Fifth International Mellor, A. M., 1981, "Semi-empirical Correlations for Gas Turbine Emissions,
Conference on Liquid Atomization and Spray Systems, Gaithersberg, MD, pp. Ignition, and Flame Stabilization," Prog. Energy Combust. Sci., Vol. 6, pp.
669-678. 347-358.
Hebrard, P., Laverge, G., Beard, P., Donnadille, P., and Trichet, P., 1993, Merryman, E. L., and Levy, A., 1975, "Nitric Oxide Formation in Flames:
"Modelisation de las Phase Liquide dans les Chambres de Combustion," AGARD The Role of NO2 and Fuel Nitrogen," Fifteenth Symposium (International) on
Conference Proceedings 536, pp. 36/1-12. Combustion, The Combustion Institute, Pittsburgh, PA, pp. 1073-1083.
Hilt, M. B., and Waslo, J., 1984, "Evolution of NO~ Abatement Techniques Micklow, G. J., Roychoudhury, S., Nguyen, H., and Cline, M. C., 1993, "Emis-
Through Combustor Design for Heavy-Duty Gas Turbines," ASME JOURNALOF sions Reduction by Varying the Swirler Airflow Split in Advanced Gas Turbine
Hung, W. S. Y., 1974, "Accurate Method of Predicting the Effect of Humidity Vol. 115, No. 3, pp. 563-569.
or Injected Water on NOx Emissions From Industrial Gas Turbines," ASME
Mongia, H. C., 1987, "A Status Report on Gas Turbine Modeling," Combus-
Paper No. 74-WA/GT-6.
tion and Fuels in Gas Tarbine Engines, AGARD CPP-422.
Hung, W. S. Y., and Agan, D. D., 1985, "The Control of NOn and CO Emissions
Mongia, H. C., 1993, "Application of CFD in Combustor Design Technology,"
From 7-MW Gas Turbines With Water Injection as Influenced by Ambient Condi-
AGARD Conference Proceedings 536, pp. 12/1-18.
tions," ASME Paper No. 85-GT-50.
Hung, W. S. Y., 1993, "Carbon Monoxide Emissions From Gas Turbines as Mosier, S. A., and Pierce, R. M., 1980, Advanced Combustion Systems for
Influenced by Ambient Temperature and Turbine Load," ASME JOURNALOF Stationa~ Gas Turbine Engines, Vol. 1, EPA Contract 68-02-2136.
ENGINEERINGFOR GAS TURBINESAND POWER, Vol. 115, No. 3, pp. 588-593. Mulhns, B. P., 1955, "Spontaneous Ignition of Liquid Fuels," AGARDograph 4.
Jasuja, A. K., and Lefebvre, A. H., 1994, "Influence of Ambient Pressure on Myers, G. D., Armstrong, J. P., White, C. D., Clouser, S., and Harvey,
Drop Size and Velocity Distributions in Dense Sprays," Twenty-Fifth Symposium R. J., 1992, "Development of an Innovative High-Temperature Gas Turbine Fuel
(International) on Combustion, The Combustion Institute, Pittsburgh, PA, pp. Nozzle," ASME JOURNALOF ENGINEERINGFOR GAS TURBINESAND POWER, Vol.
XX-00. 114, pp. 401-408.
Jones, W. P., and Priddin, C. H., 1978, "Predictions of the Flow Field and Nakata, T., Sato, M., Ninomiya, T., Yoshine, T., and Yamada, M., 1992,
Local Gas Comosition in Gas Turbine Combustors," Proceedings of the 17th "Design and Test of a Low-NOv Advanced Rich-Lean Combustor for LBG Fueled
International Symposium on Combustion, The Combustion Institute, Pittsburg, 1300°C-Class Gas Turbine," ASME Paper No. 92-GT-234.
PA, pp. 399-409. Naqwi, A., Durst, F., and Liu, X. Z., 1991, "Optical Methods for Simultaneous
Joshi, N. D., Epstein, M. J., Durlak, S., Marakovits, S., and Sabla, P. E., 1994, Measurements of Particle Size, Velocity, and Refractive Index," Applied Optics,
"Development of a Fuel-Air Premixer for Aero-Derivative Dry Low Emissions Vol. 30, No. 33, pp. 4949-4959.
Combustors," ASME Paper No. 94-GT-253. Nickolaus, D., and Lefebvre, A. H., 1987, "Fuel Thermal Stability Effects on
Kajita, S., Ohga, S., Kitajima, J., Kimura, T., and Sasaki, T., 1993, "Develop- Spray Characteristics," AIAA Journal of Propulsion and Power, Vol. 3, No. 6,
ment of a Dry Low NOn Combustor for 1.5 MW Gas Turbines," ASME Paper pp. 502-507.
No. 93-GT-393. Nicol, D. G., Malte, P. C., Lai, J., Marinov, N. N., and Pratt, D. T., 1992,
Koff, B. L., 1993, "Aircraft Gas Turbine Emissions Challenge," ASME Paper "NOx Sensitivities for Gas Turbine Engines Operated on Lean-Premixed Combus-
No. 93-GT-422. tion and Conventional Diffusion Flames," ASME Paper No. 92-GT-115.
Kumakura, H., Sasaki, M., Suzuki, D., and Ichikawa, H., 1994, "Development Nicol, D. G., Malte, P. C., and Steele, R. C., 1994, "Simplified Models for
of a Low-Emission Combustor for a 100-kW Automotive Ceramic Gas Turbine," NOw Production Rates in Lean-Premixed Combustion," ASME Paper No. 94-
ASME Paper No. 94-GT-33; accepted for publication in Trans. ASME. GT-432; accepted for publication in Trans. ASME.
Lefebvre, A. H., 1984, "Fuel Effects on Gas Turbine Combustion--Liner Novick, A. S., Troth, D. L., and Yacobucci, H. G., 1982, "Design and Prelimi-
Temperature, Pattern Factor, and Pollutant Emissions," AIAA Jom~al of Aircraft, nary Results of a Fuel Flexible Industrial Gas Turbine Combustor," ASME JOUR-
Vol. 21, No. 11, pp. 887-898. NAL OF ENGINEERINGFOR GAS TURBINESAND POWER, Vol. 104, pp. 368-376.
Lefebvre, A. H., 1986, "Fuel Effects on Gas Turbine Combustion," Interna- Odgers, J., and Kretschmer, D., 1985, "The Prediction of Thermal NOx in Gas
tional Journal of Turbo and Jet Engines, Vol. 3, No. 2, pp. 231-243. Turbines," ASME Paper No. 85-1GT-126.
Lefebvre, A. H., Freeman, W., and Cowell, L., 1986, "Spontaneous Ignition Ohkubo, Y., ldota, Y., and Nomura, Y., 1994, "Evaporation Characteristics of
Delay Characteristics of Hydrocarbon Fuel/Air Mixtures," NASA Contractor Spray in a Lean Premixed-Prevaporization Combustor for a 100 kW Automotive
Report 175064. Ceramic Gas Turbine," ASME Paper No. 94-GT-401.
Lefebvre, A. H., 1989, Atomization and Sprays, Hemisphere Publishing Corp., Ortman, J., and Lefebvre, A. H., 1985, "Fuel Distributions from Pressure-Swirl
New York. Atomizers," AIAA Journal of Propulsion and Power, Vol. 1, No. 1, pp. 11-15.
Leonard, G. L., and Correa, S. M., 1990, "NOn Formation in Lean Premixed Ozawa, Y., Fujii, T., Kikumoto, S., Sato, M., Fukuzawa, H., Saiga, M., and
High-Pressure Methane Flames," Second ASME Fossil Fuel Combustion Sympo- Watanabe, S., 1994a, "Development of Low NOx Catalytic Combustor for Gas
sium, ASME PD-30, pp. 69-74. Turbine," Proceedings of International Workshop on Catalytic Combustion,
Leonard, G., and Stegmaier, J., 1994, "Development of an Aeroderivative Gas Tokyo, Japan, Apr. 18-20, pp. 166-169.
Turbine Dry Low Emissions Combustion System," ASME JOURNAL OF ENGI- Ozawa, Y., Hirano, J., Sato, M., Saiga, M., and Watanabe, S., 1994b, "Test
NEERINGFOR GAS TURBINESAND POWER, Vol. 116, pp. 542-546. Results of Low NO~ Catalytic Combustors for Gas Turbines," ASME JOURNAL
Lewis, G. D., 1991, " A New Understanding of NO~ Formation," Proceedings OF ENGINEERINGFOR GAS TURBINESAND POWER, Wol, 116, pp. 511-516.
of International Symposium on Air-Breathing Engines, ISABE 91-7064, pp. Peng, F., and Aggarwal, S. K., 1995, " A Review of Droplet Dynamics and
625-629. Vaporization Modeling for Engineering Calculations," ASME JOURNALOF ENGI-
Lundberg, K., 1994, Ulstein Turbine Technical Report UTU94003. NEERINGFOR GAS TURBINESAND POWER, Vol. 117, July.
Lyons, V. J., 1981, "Fuel-Air Non-Uniformity Effect on Nitric Oxide Emis- Peters, J. E., 1988, "Current Gas Turbine Combustion and Fuels Research and
sions," AIAA Paper No. 81-0327. Development," AIAA Journal of Propulsion and Powet; Vol. 4, No. 3, pp.
Martin, F. J., and Dederick, P. K. J., 1976, "NOt From Fuel Nitrogen in Two- 193-206.
Stage Combustion," Sixteenth Symposium (International) on Combustion, The Pierce, R. M., Smith, C. E., and Hinton, B. S., 1980, "Advanced Combustion
Combustion Institute, Pittsburgh, PA, pp. 191-198. Systems for Stationary Gas Turbine Engines," Vol. III, EPA Contract
Maughan, J. R., Luts, A., and Bautista, P. J., 1994, ' 'A Dry Low NOx Combustor 68-02-2136.
for the MS3002 Regenerative Gas Turbine," ASME Paper No. 94-GT-252. Plee, S. L., and Mellor, A. M., 1978, "Review of Flashback Reported in
McCaldon, K., Prociw, L. A., and Sampath, P., 1993, "Design Aspects for Prevaporizing Premixing Combustors," Combustion and Flame, Vol. 32, pp.
Small Aircraft Gas Turbine Fuel Injectors," Fuels and Combustion Technology 193-203.

Journal of Engineering for Gas Turbines and Power OCTOBER 1995, Vol. 117 / 653

Downloaded From: on 01/27/2016 Terms of Use:

Poeschl, G., Ruhkamp, W., and Pfost, H., 1994, "Combustion With Low Pollut- Senior, P., Lutum, E., Polifke, W., and Sattelmayer, T., 1993, "Combustion
ant Emissions of Liquid Fuels in Gas Turbines by Premixing and Prevaporiza- Technology of the ABB GT13E2 Annular Combustor," 20th International Con-
tion," ASME Paper No. 94-GT-443. gress on Combustion Engines, CIMAC, London, Paper No. G22.
Priddin, C. H., and Coupland, J., 1986, "Impact of Numerical Methods on Serag-Eldin, M. A., and Spalding, D. B., 1979, "Computation of Three-Dimen-
Gas Turbine Combustor Design and Development," in: Calculation of Turbulent sional Gas Turbine Combustion Chamber Flows," ASME JOURNAL OF ENGI-
Reacting Flows, R. M. C. So, J. H. Whitelaw, and H. C. Mongia, eds., ASME NEERINGFOR POWER, Vol. 101, pp. 326-336.
AMD-Vol. 81, pp. 335-348. Shaw, R. J., Gilkey, S., and Hines, R., 1994, "Engine Technology Challenges
Rao, K. V. L., and Lefebvre, A. H., 1981, "Spontaneous Ignition Delay Times for a 21st Century High Speed Civil Transport," IGT1 Global Gas Turbine News,
of Hydrocarbon Fuel-Air Mixtures," First International Specialist Combustion Fall 1994, pp. 14-16.
Symposium, Bordeaux, France, pp. 325-330. Singh, R., 1994, "An Overview: Gas Turbine Generated Pollutants and the
Reeves, C. M., and Lefebvre, A. H., 1986, "Fuel Effects on Aircraft Combustor Emerging Technology Solutions," Keynote paper presented at The Gas Turbine
Emissions," ASME Paper No. 86-GT-212. Users Association 39th Annual Conference, Caracas, Venezuala, July.
Rink, K. K., and Lefebvre, A. H., 1989a, "Influence of Fuel Drop Size and Smith, K. O., Angello, L. C., and Kurzynske, F. R., 1986, "Design and Testing
Combustor Operating Conditions on Pollutant Emissions," International Journal of an Ultra-low NOx Gas Turbine Combustor," ASME Paper 86-GT-263.
of Turbo and Jet Engines, Vol. 6, No. 2, pp. 113-122. Smith, K. O., and Cowell, L. H., 1989, "Experimental Evaluation of a Liquid-
Rink, K. K., and I_.efebvre, A. H., 1989b, "The Influences of Fuel Composition Fueled, Lean-Premixed Gas Turbine Combustor," ASME Paper No. 89-GT-264.
and Spray Characteristics on Nitric Oxide Formation," Combust. Sci. and Tech., Smith, K. O., Holsapple, A. C., Mak, H. H., and Watkins, L., 1991, "Develop-
Vol. 68, pp. 1-14. ment of a Natural Gas-Fired, Ultra-Low NO~ Can Combustor for an 800 kW Gas
Rizk, N. K., and Mongia, H. C., 1986, "Gas Turbine Combustor Design Meth- Turbine," ASME Paper No. 91-GT-303.
odology," AIAA Paper No. 86-1531. Snyder, T. S., Rosfjord, T. J., McVey, J. B., and Chiappetta, L. M., 1994,
Rizk, N. K., and Lefebvre, A. H., 1987, "Prediction of Velocity Coefficient "Comparison of Liquid Fuel/Air Mixing and NOr Emissions for a Tangential
and Spray Cone Angle for Simplex Swirl Atomizers," International Journal of Entry Nozzle," ASME Paper No. 94-GT-283.
Turbo and Jet Engines, Vol. 4, pp. 65-73. Spadaccini, L. J., and Te Velde, J. A., 1980, "Autoignition Characteristics of
Rizk, N. K., Mostafa, A. A., and Mongia, H. C., 1987, "Modeling of Gas Aircraft-Type Fuels," NASA Contractor Report 159886.
Turbine Fuel Nozzles," Atomization and Spray Technology, Vol. 3, pp. Steele, R. C., Jarrett, A. C., Malte, P. C., Tonouchi, J. H., and Nicol,
241-260. D. G., 1995, "Variables Affecting NO~ Formation in Lean-Premixed Combus-
Rizk, N. K., and Mongia, H. C., 1990, "Three-Dimensional Combustor Perfor- tion," ASME Paper No. 95-GT-107; accepted for publication in Trans. ASME.
mance Validation With High-Density Fuels," AIAA Journal of Propulsion and Stickles, R. W., Dodds, W. J., Koblish, T. R., Sager, J., and Clouser, S., 1993,
Power, Vol. 6, No. 5, pp. 660-667. "Innovative High-Temperature Aircraft Engine Fuel Nozzle Design," ASME
Rizk, N. K., and Mongia, H. C., 1991a, "Low NOx Rich-Lean Combustion JOURNAL OF ENGINEERINGFOR GAS TURBINES AND POWER, Vol. 115, pp.
Concept Application," AIAA Paper No. 91-1962. 439-446.
Rizk, N. K., and Mongia, H. C., 1991b, " A Three-Dimensional Analysis of Sturgess, G. J., 1986, "Calculations of Aerospace Propulsion Combustors: A
Gas Turbine Combustors," AIAA Journal of Propulsion and Power, Vol. 7, No. View From Industry," Calculation of Turbulent Reactive Flows, R. M. C. So,
3, pp. 445-451. J. H. Whitelaw, and H. C. Mongia, eds., ASME AMD-Vol. 81, pp. 185-231.
Rizk, N. K., and Mongia, H. C., 1991c, "Gas Turbine Combustor Performance Sturgess, G. J., McKinney, R., and Morford, S. A., 1993, "Modification of
Evaluation," AIAA Paper No. 91-0640. Combustor Stoichiometry Distribution for Reduced NOx Emission From Aircraft
Rizk, N. K., and Mongia, H. C., 1992a, "Semi-analytical Fuel Injector Perfor- Engines," ASME JOURNALOF ENGINEERINGFOR GAS TURBINESAND POWER, Vol.
mance Correlation Approach," International Journal of Turbo and Jet Engines, 115, pp. 570-580.
Voh 9, pp. 205-213. Sultanian, B. K., and Mongia, H. C., 1986, "Fuel Nozzle Air Flow Modeling,"
Rizk, N. K., and Mongia, H. C., 1992b, "Calculation Approach Validation for AIAA Paper No. 86-1667.
Airblast Atomizers," ASME JOURNALOF ENGINEERINGFOR GAS TURBINESAND Tacina, R. R., 1983, "Autoignition in a Premixing-Prevaporizing Fuel Duct
POWER, Vol. 114, pp. 386-393. Using Three Different Fuel Injection Systems at Inlet Air Temperatures to
Rizk, N. K., and Mongia, H. C., 1993a, "Three-Dimensional NOv Model for 1250K," NASA TM-83938.
Rich-Lean Combustor," AIAA Paper No. 93-0251. Talpallikar, M. V., Smith, C. E., Lai, M. C., and Holdeman, J. D., 1992, "CFD
Rizk, N. K., and Mongia, H. C., 1993b, "Three-Dimensional Emission Model- Analysis of Jet Mixing in Low NOx Flametube Combustors," ASME JOURNAL
ing for Diffusion Flame, Rich-Lean, and Lean Gas Turbine Combustors," AIAA OF ENGINEERINGFOR GAS TURBINESAND POWER, Vol. 114, pp. 416-424.
Paper No. 93-2338. Tate, R. W., 1960, "Spray Patternation," Ind. Eng. Chem., Vol. 52, No. 10,
Rizk, N. K., and Mongia, H. C., 1993c, "Three-Dimensional Gas Turbine pp. 49-52.
Combustor Emissions Modeling," ASME JOURNAL OF ENGINEERINGFOR GAS Valk, M., 1994, Private Communication.
TURBINES AND POWER, Vol. 115, No. 3, pp. 603-611. Vortmeyer, N., Valk, M., and Kappler, G., 1994, " A Catalytic Combustor for
Rizk, N. K., and Mongia, H. C., 1993d, "Semianalytical Correlations for NOx, High-Temperature Gas Turbines," ASME Paper No. 94-GT-211; accepted for
AND POWER, Vol. 115, No. 3, pp. 612-619. Wang, H. Y., McDonell, V. G., and Samuelson, G. S., 1992a, "The Two-Phase
Rizk, N. K., and Mongia, H. C., 1994, "Emissions Predictions of Different Flow Downstream of a Production Engine Combustor Swirl Cup," Twenty-Fourth
Gas Turbine Combustors," AIAA Paper No. 94-0118. Symposium (International) on Combustion, The Combustion Institute, Pittsburgh,
Roberts, P. B., Kubasco, A. J., and Sekas, N. J., 1982, "Development of a PA, pp. 1457-1463.
Low NOx Lean Premixed Annular Combustor," AS ME JOURNALOF ENGINEERING Wang, H. Y., McDonell, V. G., Sowa, W. A., and Samuelson, G. S., 1992b,
FOR POWER, Vol. 104, pp. 28-35. "Scaling of the Two-Phase Flow Downstream of a Gas Turbine Combustor Swirl
Rokke, N. A., Hustad, J. E., and Berg, S., 1993, "Pollutant Emissions From Cup," ASME JOURNALOF ENGINEERINGFOR GAS TURBINES AND POWER, Wol.
Gas Fired Turbine Engines in Offshore Practice--Measurements and Scaling," 115, pp. 453-460.
ASME Paper No. 93-GT-170. Wang, H. Y., McDonell, V. G., Sowa, W. A., and Samuelson, G. S., 1994,
Rosin, P., and Rammler, E., 1933, "The Laws Governing the Fineness of "Experimental Study of a Model Gas Turbine Combustor Swirl Cup, Part II:
Powdered Coal," J. Inst. Fuel, Vol. 7, No. 31, pp. 29-36. Droplet Dynamics," AIAA Journal of Propulsion and Power.
Santangello, P. J., and Sojka, P. E., 1994, "Focused-Image Holography as a Wang, H. Y., McDonell, V. G., and Samuelson, G. S., 1995, "Influence of
Dense Spray Diagnostic," Applied Optics, Vol. 93, No. 19, pp. 4132-4136. Hardware Design on the Flow Field Structures and the Patterns of Droplet Disper-
Santavicca, D. A., Steinberger, D. L., Gibbons, K. A., Citeno, J. V., and Mills, sion: Part I--Mean Quantities," ASME JOURNALOF ENGINEERINGFOR GAS TUR-
S., 1993, "The Effect of Incomplete Fuel-Air Mixing on the Lean Limit and BINES AND POWER, Vol. 117, pp. 282-289.
Emissions Characteristics of a Lean Prevaporized Premixed (LPP) Combustor," White, D. J., Batakis, A , Le Cren, R. T., and Yacabucci, H. G., 1982, "Low
AGARD Conference Proceedings 536, pp. 22/1-12. NO~ Combustion Systems for Burning Heavy Residual Fuels and High Fuel-
Sasaki, M., Kumakura, H., and Suzuki, D., 1991, "Low NOt Combustor for Bound Nitrogen Fuels," ASME JOURNALOF ENGINEERINGFOR GAS TURBINES
Automotive Ceramic Gas Turbine--Conceptual Design," ASME Paper No. 91- AND POWER, Vol. 104, pp. 377-385.
GT-369. Wilkes, C., 1980, "Residual Fuel Combustion in Industrial Gas Turbines,"
Sattelmayer, Th., Felchlin, M. P., Haumann, J., Hellat, J., and Styner, D., 1992, in: A. H. Lefebvre, ed., Gas Turbine Combustor Design Problems, Hemisphere
"Second Generation Low-Emissions Combustors for ABB Gas Turbines: Burner Publications, New York.
Development and Tests at Atmospheric Pressure," ASME JOURNAL OF ENGI- Wilkes, C., and Gerhold, B., 1980. "NOx Reduction From a Gas Turbine Using
NEERINGFOR GAS TURBINESAND POWER, Vol. 114, No. 1, pp. 118-125. Exhaust Gas Recirculation," ASME Paper No. 80-JPGC/GT-5.
Seaton, A., MacNee, W., Donaldson, K., and Godden, E., 1995, "Particulate Willis, J. D., Teen, I. J., Schweiger, T., and Owen, D. A., 1993, "Industrial
Air Pollution and Acute Health Effects," Lancet, Vol. 345, pp. 176-178. RB211 Dry Low Emission Combustion System," ASME Paper No. 93-GT-391.
Segalman, I., McKinney, R. G., Sturgess, G. J., and Huang, L. M., 1993, Wolfer, H. H., 1938, "Der Zundverzug im Dieselmotor," V.D.L Forschungsh,
"Reduction of NOx by Fuel-Staging in Gas Turbine Engines--A Commitment Vol. 392, pp. 15-24.
to the Future," AGARD Conference Proceedings 536, pp. 29/1-17. Wynne, A., and Jasuja, A., 1991, "Visualization of Airblast Atomizer Spray
Semerjian, H. G., Ball, I. C., and Vranos, A., 1979, "Pollutant Emissions From Structure Under Varying Air Pressure Conditions," AIAA Paper No. 91-2199.
Partially Mixed Turbulent Flames," Seventeenth Symposium (International) on Zelina, J., and Ballal, D. R., 1994, "Combustion Studies in a Well-Stirred
Combustion, The Combustion Institute, Pittsburgh, PA, pp. 679-687. Reactor," AIAA Paper No. 94-0114.

654 / Vol. 117, OCTOBER 1995 Transactions of the ASME

Downloaded From: on 01/27/2016 Terms of Use: